Search Results

Search found 1884 results on 76 pages for 'backend'.

Page 68/76 | < Previous Page | 64 65 66 67 68 69 70 71 72 73 74 75  | Next Page >

• Proxying webmin with nginx

  - by TheLQ

  I am attempting to proxy webmin behind nginx for various reasons that are outside the scope of this question. However I've been trying for a while now and can't seem to figure it out and think I'm to the point where I've exhausted all the permutations of the config file I can think of. What I have now: relevant nginx config (commented out options removed, I tried many) # Proxy for webmin location /admin/quackwall-webmin { proxy_pass http://127.0.0.1:10000; # Also tried ending with /admin/quackwall-webmin proxy_set_header Host $host; } /etc/webmin/config - Relevant parts webprefix=/admin/quackwall-webmin webprefixnoredir=1 referer=(nginx domain name) Webmin itself is on the standard ports, listening on all addresses temporarily for debugging. SSL has been disabled for right now. So I make a standard request for the login page. However all the CSS and images are broken, with the standard login page returned for all of the resources. In the webmin miniserv logs I see 127.0.0.1 - - [29/Oct/2012:12:29:00 -0400] "GET /admin/quackwall-webmin/session_login.cgi HTTP/1.0" 401 2453 127.0.0.1 - - [29/Oct/2012:12:29:01 -0400] "GET /admin/quackwall-webmin/unauthenticated/style.css HTTP/1.0" 401 2453 127.0.0.1 - - [29/Oct/2012:12:29:01 -0400] "GET /admin/quackwall-webmin/unauthenticated/sorttable.js HTTP/1.0" 401 2453 127.0.0.1 - - [29/Oct/2012:12:29:01 -0400] "GET /admin/quackwall-webmin/unauthenticated/toggleview.js HTTP/1.0" 401 2453 So all the URL's are returning 401s. Interestingly ngrep seems to show that the requests suceeded on the backend communication between nginx and webmin T 127.0.0.1:58908 -> 127.0.0.1:10000 [AP] POST /admin/quackwall-webmin/session_login.cgi HTTP/1.0..Host: (host)..Connection: close..User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW 64; rv:16.0) Gecko/20100101 Firefox/16.0..Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8..Accept-Language: en-US,en;q=0.5. .Accept-Encoding: gzip, deflate..Referer: http://(host)/admin/quackwall-webmin/session_login.cgi..Cookie: testing=1..Cache-Control: ma x-age=0..Content-Type: application/x-www-form-urlencoded..Content-Length: 41....page=%2F&user=(user)&pass=(pass) T 127.0.0.1:10000 -> 127.0.0.1:58908 [AP] HTTP/1.0 200 Document follows.. Various other permutations of these config options and others show similar results, with the URL sent to webmin by nginx either being /admin/quackwall-webmin/session_login.cgi, /admin/quackwall-webmin//session_login.cgi, and just /session_login.cgi. All give 201 Unauthenticated responses. All requests, even those that somewhat succeed (as in I can actually load the resources of the page) Is changing the webprefix in webmin even supported? What am I doing wrong? What else can I try?

  Read the article

• OS X: Finder error -36 when using SMB shares on a Samba server bound to AD

  - by Frenchie

  We're looking at deploying SMB homes on Debian (5.0.3) for our mac clients rather than purchasing four new Xserves. We've got our test servers built and functioning properly. Windows clients behave perfectly, but we've run into an issue with OS X (10.6.x and 10.5.x). We're going this route instead of Windows file servers due to a whole bunch of other issues that arise when going that way. Specifically, when mounting a SMB share with unix extensions switched on and the remote server bound to AD, the finder cannot save files on the share, instead touching the file and then bombing out with a -36 IO error, folder creation is fine. Copying files in the terminal behaves fine and the problem seems to be limited to the finder. The issue arises (I think) as the remote UID/GID is passed across when using unix extensions. OS X uses its own winbind idmap (odsam) to work out the effective UID/GID from AD users and groups whilst we're using a rid map on the server. Consequently, there is a mismatch in ownership which the finder chooses to honour. How OS X appears to handle this is to use the remote uid and gid at the file permission level (see below) and then set an OS X acl granting the local uid/gid to have the appropriate permissions on the file. I think the finder touches the file (which the kernel allows because of the ACL) and then checks the filesystem perms and drops out with the IO error. On a Client fc-003353-d:homes2 root# ls -led test/ drwx------+ 2 135978 100513 16384 Feb 3 15:14 test/ 0: user:jfrench allow list,add_file,search,delete,add_subdirectory,delete_child,readattr,writeattr,readextattr,writeextattr,readsecurity,writesecurity,chown,file_inherit,directory_inherit 1: group:ARTS\domain users allow 2: group:everyone allow 3: group:owner allow list,add_file,search,delete,add_subdirectory,delete_child,readattr,writeattr,readextattr,writeextattr,readsecurity,writesecurity,chown,file_inherit,directory_inherit,only_inherit 4: group:group allow list,add_file,search,delete,add_subdirectory,delete_child,readattr,writeattr,readextattr,writeextattr,readsecurity,writesecurity,chown,file_inherit,directory_inherit,only_inherit 5: group:everyone allow list,add_file,search,delete,add_subdirectory,delete_child,readattr,writeattr,readextattr,writeextattr,readsecurity,writesecurity,chown,file_inherit,directory_inherit,only_inherit We've tried the following without any luck: Setting the Linux side file owner to match the OS X GID/UID Adding ACLs on the linux filesystem which grant the OS X GID/UID perms Disabling extended attributes Setting steams=no in /etc/nsmb.conf on the client We're currently running a workaround which is to just turn off unix extensions which forces the macs to just mount the share as the local user with u=rwx perms. This works for most things but is causing a few apps that expect certain perms to break in subtle ways. Worst case scenario is that we'll continue running in this way but we would like to have the unix extensions on. Regards. Relevant SMB config below: [global] workgroup = ARTS realm = *snip* security = ADS password server = *snip* unix extensions = yes panic action = /usr/share/panic-action %d idmap backend = rid:ARTS=100000-10000000 idmap uid = 100000-10000000 idmap gid = 100000-10000000 winbind enum users = Yes winbind enum groups = Yes veto files = /lost+found/aquota.*/ hide files = /desktop.ini/$RECYCLE.BIN/.*/AppData/Library/ ea support = yes store dos attributes = yes map system = no map archive = no map readonly = no

  Read the article

• Configuring OpenLDAP as a Active Directory Proxy

  - by vadensumbra

  We try to set up an Active Directory server for company-wide authentication. Some of the servers that should authenticate against the AD are placed in a DMZ, so we thought of using a LDAP-server as a proxy, so that only 1 server in the DMZ has to connect to the LAN where the AD-server is placed). With some googling it was no problem to configure the slapd (see slapd.conf below) and it seemed to work when using the ldapsearch tool, so we tried to use it in apache2 htaccess to authenticate the user over the LDAP-proxy. And here comes the problem: We found out the username in the AD is stored in the attribute 'sAMAccountName' so we configured it in .htaccess (see below) but the login didn't work. In the syslog we found out that the filter for the ldapsearch was not (like it should be) '(&(objectClass=*)(sAMAccountName=authtest01))' but '(&(objectClass=*)(?=undefined))' which we found out is slapd's way to show that the attribute do not exists or the value is syntactically wrong for this attribute. We thought of a missing schema and found the microsoft.schema (and the .std / .ext ones of it) and tried to include them in the slapd.conf. Which does not work. We found no working schemata so we just picked out the part about the sAMAccountName and build a microsoft.minimal.schema (see below) that we included. Now we get the more precise log in the syslog: Jun 16 13:32:04 breauthsrv01 slapd[21229]: get_ava: illegal value for attributeType sAMAccountName Jun 16 13:32:04 breauthsrv01 slapd[21229]: conn=0 op=1 SRCH base="ou=oraise,dc=int,dc=oraise,dc=de" scope=2 deref=3 filter="(&(objectClass=\*)(?sAMAccountName=authtest01))" Jun 16 13:32:04 breauthsrv01 slapd[21229]: conn=0 op=1 SRCH attr=sAMAccountName Jun 16 13:32:04 breauthsrv01 slapd[21229]: conn=0 op=1 SEARCH RESULT tag=101 err=0 nentries=0 text= Using our Apache htaccess directly with the AD via LDAP works though. Anyone got a working setup? Thanks for any help in advance: slapd.conf: allow bind_v2 include /etc/ldap/schema/core.schema ... include /etc/ldap/schema/microsoft.minimal.schema ... backend ldap database ldap suffix "ou=xxx,dc=int,dc=xxx,dc=de" uri "ldap://80.156.177.161:389" acl-bind bindmethod=simple binddn="CN=authtest01,ou=GPO-Test,ou=xxx,dc=int,dc=xxx,dc=de" credentials=xxxxx .htaccess: AuthBasicProvider ldap AuthType basic AuthName "AuthTest" AuthLDAPURL "ldap://breauthsrv01.xxx.de:389/OU=xxx,DC=int,DC=xxx,DC=de?sAMAccountName?sub" AuthzLDAPAuthoritative On AuthLDAPGroupAttribute member AuthLDAPBindDN CN=authtest02,OU=GPO-Test,OU=xxx,DC=int,DC=xxx,DC=de AuthLDAPBindPassword test123 Require valid-user microsoft.minimal.schema: attributetype ( 1.2.840.113556.1.4.221 NAME 'sAMAccountName' SYNTAX '1.3.6.1.4.1.1466.115.121.1.15' SINGLE-VALUE )

  Read the article

• Configure Postfix to Port other than 25

  - by bwheeler96

  I've done quite a bit of googling on how to reconfigure postfix to work on a different port, but I still can't fond the line(s) people keep talking about in my master.cf. I'm using OS X Mountain Lion, and my ISP blocks traffic both ways on port 25. people have said to look for a line that says smtp inet n - n - - smtpd I can't find it. This is (what I believe to be) unmodified # ==== Begin auto-generated section ======================================== # This section of the master.cf file is auto-generated by the Server Admin # Mail backend plugin whenever mails settings are modified. smtp inet n - n - 1 postscreen smtpd pass - - n - - smtpd dnsblog unix - - n - 0 dnsblog tlsproxy unix - - n - 0 tlsproxy submission inet n - n - - smtpd -o smtpd_tls_security_level=encrypt smtp unix - - n - - smtp # === End auto-generated section =========================================== # Modern SMTP clients communicate securely over port 25 using the STARTTLS command. # Some older clients, such as Outlook 2000 and its predecessors, do not properly # support this command and instead assume a preconfigured secure connection # on port 465. This was sometimes called "smtps", but such usage was never # approved by the IANA and therefore conflicts with another, legitimate assignment. # For more details about managing secure SMTP connections with postfix, please see: # http://www.postfix.org/TLS_README.html # To read more about configuring secure connections with Outlook 2000, please read: # http://support.microsoft.com/default.aspx?scid=kb;en-us;Q307772 # Apple does not support the use of port 465 for this purpose. # After determining that connecting clients do require this behavior, you may choose # to manually enable support for these older clients by uncommenting the following # four lines. #465 inet n - n - - smtpd # -o smtpd_tls_wrappermode=yes # -o smtpd_sasl_auth_enable=yes # -o smtpd_client_restrictions=permit_sasl_authenticated,reject # -o milter_macro_daemon_name=ORIGINATING #628 inet n - n - - smtp pickup fifo n - n 60 1 pickup cleanup unix n - n - 0 cleanup qmgr fifo n - n 300 1 qmgr #qmgr fifo n - n 300 1 oqmgr tlsmgr unix - - n 1000? 1 tlsmgr rewrite unix - - n - - trivial-rewrite bounce unix - - n - 0 bounce defer unix - - n - 0 bounce trace unix - - n - 0 bounce verify unix - - n - 1 verify sacl-cache unix - - n - 1 sacl-cache flush unix n - n 1000? 0 flush proxymap unix - - n - - proxymap proxywrite unix - - n - 1 proxymap # When relaying mail as backup MX, disable fallback_relay to avoid MX loops relay unix - - n - - smtp -o smtp_fallback_relay= # -o smtp_helo_timeout=5 -o smtp_connect_timeout=5 showq unix n - n - - showq error unix - - n - - error retry unix - - n - - error discard unix - - n - - discard local unix - n n - - local virtual unix - n n - - virtual lmtp unix - - n - - lmtp anvil unix - - n - 1 anvil scache unix - - n - 1 scache # # ==================================================================== # Interfaces to non-Postfix software. Be sure to examine the manual # pages of the non-Postfix software to find out what options it wants.

  Read the article

• Varnish configuration to only cache for non-logged in users

  - by davidsmalley

  I have a Ruby on Rails application fronted by varnish+nginx. As most of the sites content is static unless you are a logged in user, I want to cache the site heavily with varnish when a user is logged out but only to cache static assets when they are logged in. When a user is logged in they will have the cookie 'user_credentials' present in their Cookie: header, in addition I need to skip caching on /login and /sessions in order that a user can get their 'user_credentials' cookie in the first place. Rails by default does not set a cache friendly Cache-control header, but my application sets a "public,s-max-age=60" header when a user is not logged in. Nginx is set to return 'far future' expires headers for all static assets. The configuration I have at the moment is totally bypassing the cache for everything when logged in, including static assets — and is returning cache MISS for everything when logged out. I've spent hours going around in circles and here is my current default.vcl director rails_director round-robin { { .backend = { .host = "xxx.xxx.xxx.xxx"; .port = "http"; .probe = { .url = "/lbcheck/lbuptest"; .timeout = 0.3 s; .window = 8; .threshold = 3; } } } } sub vcl_recv { if (req.url ~ "^/login") { pipe; } if (req.url ~ "^/sessions") { pipe; } # The regex used here matches the standard rails cache buster urls # e.g. /images/an-image.png?1234567 if (req.url ~ "\.(css|js|jpg|jpeg|gif|ico|png)\??\d*$") { unset req.http.cookie; lookup; } else { if (req.http.cookie ~ "user_credentials") { pipe; } } # Only cache GET and HEAD requests if (req.request != "GET" && req.request != "HEAD") { pipe; } } sub vcl_fetch { if (req.url ~ "^/login") { pass; } if (req.url ~ "^/sessions") { pass; } if (req.http.cookie ~ "user_credentials") { pass; } else { unset req.http.Set-Cookie; } # cache CSS and JS files if (req.url ~ "\.(css|js|jpg|jpeg|gif|ico|png)\??\d*$") { unset req.http.Set-Cookie; } if (obj.status >=400 && obj.status <500) { error 404 "File not found"; } if (obj.status >=500 && obj.status <600) { error 503 "File is Temporarily Unavailable"; } } sub vcl_deliver { if (obj.hits > 0) { set resp.http.X-Cache = "HIT"; } else { set resp.http.X-Cache = "MISS"; } }

  Read the article

• nginx proxypath https redirect fails without trailing slash

  - by Thermionix

  I'm trying to setup Nginx to forward requests to several backend services using proxy_pass. The links on the pages that lack trailing slashes do have https:// in front, but get redirected to a http request with a trailing slash - which ends in connection refused - I only want these services to be available through https. So if a link is too https://example.com/internal/errorlogs in a browser when loaded https://example.com/internal/errorlogs gives Error Code 10061: Connection refused (it redirects to http://example.com/internal/errorlogs/) If I manually append the trialing slash https://example.com/internal/errorlogs/ it loads I've tried with varied trailing forward slashes appended to the proxypath and location in proxy.conf to no effect, have also added server_name_in_redirect off; This happens on more than one app under nginx, and works in apache reverse proxy Config files; proxy.conf location /internal { proxy_pass http://localhost:8081/internal; include proxy.inc; } .... more entries .... sites-enabled/main server { listen 443; server_name example.com; server_name_in_redirect off; include proxy.conf; ssl on; } proxy.inc proxy_connect_timeout 59s; proxy_send_timeout 600; proxy_read_timeout 600; proxy_buffer_size 64k; proxy_buffers 16 32k; proxy_pass_header Set-Cookie; proxy_redirect off; proxy_hide_header Vary; proxy_busy_buffers_size 64k; proxy_temp_file_write_size 64k; proxy_set_header Accept-Encoding ''; proxy_ignore_headers Cache-Control Expires; proxy_set_header Referer $http_referer; proxy_set_header Host $host; proxy_set_header Cookie $http_cookie; proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-Host $host; proxy_set_header X-Forwarded-Server $host; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header X-Forwarded-Ssl on; proxy_set_header X-Forwarded-Proto https; curl output -$ curl -I -k https://example.com/internal/errorlogs/ HTTP/1.1 200 OK Server: nginx/1.0.5 Date: Thu, 24 Nov 2011 23:32:07 GMT Content-Type: text/html;charset=utf-8 Connection: keep-alive Content-Length: 14327 -$ curl -I -k https://example.com/internal/errorlogs HTTP/1.1 301 Moved Permanently Server: nginx/1.0.5 Date: Thu, 24 Nov 2011 23:32:11 GMT Content-Type: text/html;charset=utf-8 Connection: keep-alive Content-Length: 127 Location: http://example.com/internal/errorlogs/

  Read the article

• Nginx Retry of Requests ( Nginx - Haproxy Combination )

  - by vaibhav

  I wanted to ask about Nginx Retry of Requests. I have a Nginx running at the backend which then sends the requests to HaProxy which then passes it on the web server and the request is processed. I am reloading my Haproxy config dynamically to provide elasticity. The problem is that the requests are dropped when I reload Haproxy. So I wanted to have a solution where I can just retry that from Nginx. I looked through the proxy_connect_timeout, proxy_next_upstream in http module and max_fails and fail_timeout in server module. I initially only had 1 server in the upstream connections so I just that up twice now and less requests are getting dropped ( only when ) have say the same server twice in upstream , if I have same server 3-4 times drops increase ). So , firstly I wanted to now , that when a request is not able to establish connection from Nginx to Haproxy so while reloading it seems that conneciton is seen as error and straightway the request is dropped . So how can I either specify the time after the failure I want to retry the request from Nginx to upstream or the time before which Nginx treats it as failed request. ( I have tried increaing proxy_connect_timeout - didn't help , mail_retires , fail_timeout and also putting the same upstream server twice ( that gave the best results so far ) Nginx Conf File upstream gae_sleep { server 128.111.55.219:10000; } server { listen 8080; server_name 128.111.55.219; root /var/apps/sleep/app; # Uncomment these lines to enable logging, and comment out the following two #access_log /var/log/nginx/sleep.access.log upstream; error_log /var/log/nginx/sleep.error.log; access_log off; #error_log /dev/null crit; rewrite_log off; error_page 404 = /404.html; set $cache_dir /var/apps/sleep/cache; location / { proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header Host $http_host; proxy_redirect off; proxy_pass http://gae_sleep; client_max_body_size 2G; proxy_connect_timeout 30; client_body_timeout 30; proxy_read_timeout 30; } location /404.html { root /var/apps/sleep; } location /reserved-channel-appscale-path { proxy_buffering off; tcp_nodelay on; keepalive_timeout 55; proxy_pass http://128.111.55.219:5280/http-bind; } }

  Read the article

• nginx reverse proxy subdomain is redirecting

  - by holtkampw

  So I have a frontend nginx server which will proxy to several other nginx servers (running Passenger for Rails apps). Here's the part of the frontend nginx config in question: server { listen 80; server_name git.domain.com; access_log /server/domain/log/nginx.access.log; error_log /server/domain/log/nginx_error.log debug; location / { proxy_pass http://127.0.0.1:8020/; proxy_redirect off; proxy_set_header Host $host; proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_max_temp_file_size 0; client_max_body_size 10m; client_body_buffer_size 128k; proxy_connect_timeout 90; proxy_send_timeout 90; proxy_read_timeout 90; proxy_buffer_size 4k; proxy_buffers 4 32k; proxy_busy_buffers_size 64k; proxy_temp_file_write_size 64k; } } server { listen 80; server_name domain.com; access_log /server/domain/log/nginx.access.log; error_log /server/domain/log/nginx_error.log debug; location / { proxy_pass http://127.0.0.1:8000/; proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header Host $http_host; proxy_redirect off; proxy_set_header X_FORWARDED_PROTO https; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; } } Finally here's the backend for git.domain.com: server { listen 8020; #server_name localhost; root /server/gitorious/gitorious/public/; passenger_enabled on; proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header Host $http_host; proxy_redirect off; proxy_set_header X_FORWARDED_PROTO https; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; } So here's the problem. When I type in git.domain.com, my gitorious install will redirect to domain.com. It works perfect there, but it ignores the subdomain. At first I thought it was the server_name construct. I have tried git.domain.com, domain.com, localhost, and currently none. Any ideas?

  Read the article

• Configuring varnish and django (apache/modwsgi)

  - by Hedde

  I am trying to work out why my application keeps hitting the database while I have setup varnish infront of apache. I think I am missing some vital configuration, any tips are welcome This is my curl result: HTTP/1.1 200 OK Server: Apache/2.2.16 (Debian) Content-Language: en-us Vary: Accept,Accept-Encoding,Accept-Language,Cookie Cache-Control: s-maxage=60, no-transform, max-age=60 Content-Type: application/json; charset=utf-8 Date: Sat, 15 Sep 2012 08:19:17 GMT Connection: keep-alive My varnishlog: 13 BackendClose - apache 13 BackendOpen b apache 127.0.0.1 47665 127.0.0.1 8000 13 TxRequest b GET 13 TxURL b /api/v1/events/?format=json 13 TxProtocol b HTTP/1.1 13 TxHeader b User-Agent: curl/7.19.7 (universal-apple-darwin10.0) libcurl/7.19.7 OpenSSL/0.9.8r zlib/1.2.3 13 TxHeader b Host: foobar.com 13 TxHeader b Accept: */* 13 TxHeader b X-Forwarded-For: 92.64.200.145 13 TxHeader b X-Varnish: 979305817 13 TxHeader b Accept-Encoding: gzip 13 RxProtocol b HTTP/1.1 13 RxStatus b 200 13 RxResponse b OK 13 RxHeader b Date: Sat, 15 Sep 2012 08:21:28 GMT 13 RxHeader b Server: Apache/2.2.16 (Debian) 13 RxHeader b Content-Language: en-us 13 RxHeader b Content-Encoding: gzip 13 RxHeader b Vary: Accept,Accept-Encoding,Accept-Language,Cookie 13 RxHeader b Cache-Control: s-maxage=60, no-transform, max-age=60 13 RxHeader b Content-Length: 6399 13 RxHeader b Content-Type: application/json; charset=utf-8 13 Fetch_Body b 4(length) cls 0 mklen 1 13 Length b 6399 13 BackendReuse b apache 11 SessionOpen c 92.64.200.145 53236 :80 11 ReqStart c 92.64.200.145 53236 979305817 11 RxRequest c HEAD 11 RxURL c /api/v1/events/?format=json 11 RxProtocol c HTTP/1.1 11 RxHeader c User-Agent: curl/7.19.7 (universal-apple-darwin10.0) libcurl/7.19.7 OpenSSL/0.9.8r zlib/1.2.3 11 RxHeader c Host: foobar.com 11 RxHeader c Accept: */* 11 VCL_call c recv lookup 11 VCL_call c hash 11 Hash c /api/v1/events/?format=json 11 Hash c foobar.com 11 VCL_return c hash 11 VCL_call c miss fetch 11 Backend c 13 apache apache 11 TTL c 979305817 RFC 60 -1 -1 1347697289 0 1347697288 0 60 11 VCL_call c fetch deliver 11 ObjProtocol c HTTP/1.1 11 ObjResponse c OK 11 ObjHeader c Date: Sat, 15 Sep 2012 08:21:28 GMT 11 ObjHeader c Server: Apache/2.2.16 (Debian) 11 ObjHeader c Content-Language: en-us 11 ObjHeader c Content-Encoding: gzip 11 ObjHeader c Vary: Accept,Accept-Encoding,Accept-Language,Cookie 11 ObjHeader c Cache-Control: s-maxage=60, no-transform, max-age=60 11 ObjHeader c Content-Type: application/json; charset=utf-8 11 Gzip c u F - 6399 69865 80 80 51128 11 VCL_call c deliver deliver 11 TxProtocol c HTTP/1.1 11 TxStatus c 200 11 TxResponse c OK 11 TxHeader c Server: Apache/2.2.16 (Debian) 11 TxHeader c Content-Language: en-us 11 TxHeader c Vary: Accept,Accept-Encoding,Accept-Language,Cookie 11 TxHeader c Cache-Control: s-maxage=60, no-transform, max-age=60 11 TxHeader c Content-Type: application/json; charset=utf-8 11 TxHeader c Date: Sat, 15 Sep 2012 08:21:29 GMT 11 TxHeader c Connection: keep-alive 11 Length c 0 11 ReqEnd c 979305817 1347697288.292612076 1347697289.456128597 0.000086784 1.163468122 0.000048399

  Read the article

• My server is slower than the average user's computer, should I still offload Access queries to SQL Server? [closed]

  - by andrewb

  Possible Duplicate: How do you do Load Testing and Capacity Planning for Databases I have a database set up with MS Access 2007 front ends and an SQL Server 2005 back end. At the moment, all the queries are saved in the front end as I've only recently moved to an SQL Server backend. I'm wondering how much of those queries I should save as stored procedures/views on SQL Server. About the system The number of concurrent users is only a handful, though it could be as high as 25 at one time (very unlikely). The average computer has an Intel i3-2120 CPU running at 3.3 GHz, which gets a PassMark score of 3,987, whilst the server has an Intel Xeon E5335 running at 2.0 GHz, which gets a PassMark score of 2,637. Always an awkward situation when an i3 outperforms a Xeon... though the i3 is from Q1 2011 and the Xeon is Q2 2009. There is potential for a server upgrade in the future, though it wouldn't come easy. I'm inclined to move the queries to the back end, as they are beginning to take noticeable time and I figure that is a better way of doing things. I like the idea of throwing everything at the server, then pushing for a server upgrade. It makes more sense in my mind to be upgrading one server rather than 30 PCs. Or am I being overzealous? Why my question isn't a duplicate It seems that my question has been misinterpreted and labelled a duplicate of quite a different question, one about testing and capacity planning. I'll try explain how my question is very different from the linked question. The crux of my question is something like "Even though my server is technically slower, is it better to have it doing more of the queries?" There's two ways that people could have answered this: I agree the server is going to be slower, but the extra benefits of such and such (like the less Access the better) means you should move most to the server anyway. (OR no it doesn't outweigh the benefit, keep them in Access) Actually the server will be faster because of such and such. I'm hoping that people out there could provide some answers like this, and the question in the dupe link doesn't really provide either of these answers. Ok sure, I suppose I could do extensive performance testing to compare Access queries running on a local machine to SQL Server queries running on the server, but that sounds like a very hard task (particularly performance testing of access) compared to someone giving some quick general guidance, and again, my question is looking for a lot more than immediate performance benefit.

  Read the article

• Authenticate by libpam-mysql and libnss-mysql (CentOS)

  - by Chris

  I'm trying to get MySQL to function as a backend for authenticating users on CentOS 6.3. So far I have successfully installed and configured libnss-mysql. I can test this by doing: # groups testuser testuser : sftp Testuser is a member of the sftp group in fact, all MySQL based useraccounts will be hardcoded to it. The sftp group is chrooted and forced to use internal-sftp so they cannot do anything but access their home directory. Then I configured pam-mysql and PAM to allow mysql logins. This also works.. When SELinux is not enforcing. When I do setenforce 1 users can no longer login. Error: Permission denied, please try again. This is my pam_mysql.conf file: users.host=localhost users.db_user=nss-pam-user users.db_passwd=*********** users.database=sftpusers users.table=users users.user_column=username users.password_column=password users.password_crypt=6 verbose=1 My /etc/pam.d/sshd: #%PAM-1.0 auth sufficient pam_sepermit.so auth include password-auth auth required pam_mysql.so config_file=/etc/pam_mysql.conf account sufficient pam_nologin.so account include password-auth account required pam_mysql.so config_file=/etc/pam_mysql.conf password include password-auth # pam_selinux.so close should be the first session rule session required pam_selinux.so close session required pam_loginuid.so # pam_selinux.so open should only be followed by sessions to be executed in the user context session required pam_selinux.so open env_params session optional pam_keyinit.so force revoke session include password-auth And to be complete the contents of some log files.. /var/logs/secure Nov 20 14:52:20 hostname unix_chkpwd[4891]: check pass; user unknown Nov 20 14:52:20 hostname unix_chkpwd[4891]: password check failed for user (testuser) Nov 20 14:52:20 hostname sshd[4880]: pam_unix(sshd:auth): authentication failure; logname= uid=0 euid=0 tty=ssh ruser= rhost=192.168.10.107 user=testuser Nov 20 14:52:22 sftpusers sshd[4880]: Failed password for testuser from 192.168.10.107 port 51849 ssh2 /var/logs/audit/audit.log type=USER_AUTH msg=audit(1353420107.070:812): user pid=5285 uid=0 auid=500 ses=24 subj=unconfined_u:system_r:sshd_t:s0-s0:c0.c1023 msg='op=pubkey acct="testuser" exe="/usr/sbin/sshd" hostname=? addr=192.168.10.107 terminal=ssh res=failed' type=USER_AUTH msg=audit(1353420112.312:813): user pid=5285 uid=0 auid=500 ses=24 subj=unconfined_u:system_r:sshd_t:s0-s0:c0.c1023 msg='op=PAM:authentication acct="testuser" exe="/usr/sbin/sshd" hostname=192.168.10.107 addr=192.168.10.107 terminal=ssh res=failed' type=USER_AUTH msg=audit(1353420112.456:814): user pid=5285 uid=0 auid=500 ses=24 subj=unconfined_u:system_r:sshd_t:s0-s0:c0.c1023 msg='op=password acct="testuser" exe="/usr/sbin/sshd" hostname=? addr=192.168.10.107 terminal=ssh res=failed' I tried to let audit2why explain the problem but it remains silent even though there are some errors. Does anyone see the problem? Thanks! EDIT: Turns out it's almost working with setenforce 0 I can mkdir foobar but if I do a single ls I get an error: Received message too long 16777216

  Read the article

• nginx connection time issue on some IPs

  - by sheldon

  I have recently shifted my server to nginx and php-fpm getting rid of apache. This has helped improves speeds of my website. Everything seems to work fine until i came across this issue, i noticed that nginx keeps throwing connection time out errors for only certain IPs. One of the IPs is my office IP, we have a backend that is accessed from our office through out the day. I use supervisord to launch 3 php-fpm processes with workers this is my typical php-fpm config pm.max_children = 50 pm.start_servers = 20 pm.min_spare_servers = 5 pm.max_spare_servers = 35 pm.max_requests = 300 Since i have a server with 4 cores and 2 GB ram this is my nginx setup worker_processes 4; worker_rlimit_nofile 8192; events { worker_connections 1024; use epoll; multi_accept off; } sendfile on; tcp_nopush on; tcp_nodelay on; keepalive_timeout 55; recursive_error_pages on; server_name_in_redirect off; server_tokens off; client_header_timeout 3m; client_body_timeout 3m; send_timeout 3m; connection_pool_size 256; client_header_buffer_size 8k; large_client_header_buffers 4 32k; request_pool_size 4k; output_buffers 4 32k; postpone_output 1460; proxy_buffer_size 32k; proxy_buffers 4 32k; proxy_busy_buffers_size 64k; proxy_temp_file_write_size 64k; fastcgi_connect_timeout 120; fastcgi_send_timeout 120; fastcgi_read_timeout 180; fastcgi_buffer_size 128k; fastcgi_buffers 4 256k; fastcgi_busy_buffers_size 256k; fastcgi_temp_file_write_size 256k; fastcgi_intercept_errors on; fastcgi_ignore_client_abort off; Where am i going wrong with the config, I have tried various settings but the issue still persists. These are the errors i keep getting 2011/11/13 18:20:33 [error] 21583#0: *311683 upstream timed out (110: Connection timed out) while reading response header from upstream, client: IP, server: tastykhana.in, request: "GET url HTTP/1.1", upstream: "fastcgi://unix:/var/run/php-fpm.socket:", host: "tastykhana.in", referrer: "url"

  Read the article

• How to make Horde connect to mysql with UTF-8 character set?

  - by jkj

  How to tell horde 3.3.11 to use UTF-8 for it's mysql connection? The $conf['sql']['charset'] only tells horde what is expected from the database. Horde uses MDB2 to connect to mysql. Is there way to force MDB2 or mysql character_set_client from php.ini? So far I found two workarounds: Force mysql to ignore character set requested by client [mysqld] skip-character-set-client-handshake=1 default-character-set=utf8 Force mysql to run SET NAMES utf8 on every connection [mysqld] init-connect='SET NAMES utf8' Both have drawbacks on multi user mysql server. The first disables converting character sets alltogether and the second one forces every connection to produce UTF-8. [EDIT] Found the problem. The 'charset' parameter was unset the last minute before sending to SQL backend. This is probably due to mysql not being able to digest utf-8 but utf8. Mysql specific mapping is required to make it work. I just worked around it by translating utf-8 - utf8. Won't work with any other databases with this patch though. --- lib/Horde/Share/sql.php.orig 2011-07-04 17:09:33.349334890 +0300 +++ lib/Horde/Share/sql.php 2011-07-04 17:11:06.238636462 +0300 @@ -753,7 +753,13 @@ /* Connect to the sql server using the supplied parameters. */ require_once 'MDB2.php'; $params = $this->_params; - unset($params['charset']); + + if ($params['charset'] == 'utf-8') { + $params['charset'] = 'utf8'; + } else { + unset($params['charset']); + } + $this->_write_db = &MDB2::factory($params); if (is_a($this->_write_db, 'PEAR_Error')) { Horde::fatal($this->_write_db, __FILE__, __LINE__); @@ -792,7 +798,13 @@ /* Check if we need to set up the read DB connection seperately. */ if (!empty($this->_params['splitread'])) { $params = array_merge($params, $this->_params['read']); - unset($params['charset']); + + if ($params['charset'] == 'utf-8') { + $params['charset'] = 'utf8'; + } else { + unset($params['charset']); + } + $this->_db = &MDB2::singleton($params); if (is_a($this->_db, 'PEAR_Error')) { Horde::fatal($this->_db, __FILE__, __LINE__);

  Read the article

• nginx proxypath https redirects to http

  - by Thermionix

  I'm trying to setup Nginx to forward requests to several backend services using proxy_pass however several pages load with 404s The links on the pages have https:// in front, but result in a http request - which ends in a 404 - I only want these services to be available through https. I've tried with varied trailing forward slashes appended to the proxypath and location in proxy.conf, I've also tried commenting out www.conf (just incase its location blocks could have caused any conflicts) to no effect. So if a link is too https://example.com/sickbeard/errorlogs in a browser when loaded https://example.com/sickbeard/errorlogs gives a 404 in a browser https://example.com/sickbeard/errorlogs/ loads nginx error log; 2011/11/23 14:21:58 [error] 28882#0: *6 "/var/www/sickbeard/errorlogs/recent.html" is not found (2: No such file or directory), client: 192.168.1.99, server: example.com, request: "GET /sickbeard/errorlogs/ HTTP/1.1", host: "example.com" Config files; proxy.conf location /sickbeard { proxy_pass http://localhost:8081/sickbeard; include proxy.inc; } .... more entries .... sites-enabled/main server { listen 80; include www.conf; } server { listen 443; include proxy.conf; include www.conf; ssl on; } www.conf root /var/www; server_name example.com; location / { autoindex off; allow all; rewrite ^/$ /mainsite last; location ~* \.(jpg|jpeg|gif|css|png|js|ico)$ { expires max; } location ~ \.php$ { fastcgi_index index.php; include fastcgi_params; if (-f $request_filename) { fastcgi_pass 127.0.0.1:9000; } } } proxy.inc proxy_connect_timeout 59s; proxy_send_timeout 600; proxy_read_timeout 600; proxy_buffer_size 64k; proxy_buffers 16 32k; proxy_pass_header Set-Cookie; proxy_redirect off; proxy_hide_header Vary; proxy_busy_buffers_size 64k; proxy_temp_file_write_size 64k; proxy_set_header Accept-Encoding ''; proxy_ignore_headers Cache-Control Expires; proxy_set_header Referer $http_referer; proxy_set_header Host $host; proxy_set_header Cookie $http_cookie; proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-Host $host; proxy_set_header X-Forwarded-Server $host; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;

  Read the article

• How to keep group-writeable shares on Samba with OSX clients?

  - by Oliver Salzburg

  I have a FreeNAS server on a network with OSX and Windows clients. When the OSX clients interact with SMB/CIFS shares on the server, they are causing permission problems for all other clients. Update: I can no longer verify any answers because we abandoned the project, but feel free to post any help for future visitors. The details of this behavior seem to also be dependent on the version of OSX the client is running. For this question, let's assume a client running 10.8.2. When I mount the CIFS share on an OSX client and create a new directory on it, the directory will be created with drwxr-x-rx permissions. This is undesirable because it will not allow anyone but me to write to the directory. There are other users in my group which should have write permissions as well. This behavior happens even though the following settings are present in smb.conf on the server: [global] create mask= 0666 directory mask= 0777 [share] force directory mode= 0775 force create mode= 0660 I was under the impression that these settings should make sure that directories are at least created with rwxrwxr-x permissions. But, I guess, that doesn't stop the client from changing the permissions after creating the directory. When I create a folder on the same share from a Windows client, the new folder will have the desired access permissions (rwxrwxrwx), so I'm currently assuming that the problem lies with the OSX client. I guess this wouldn't be such an issue if you could easily change the permissions of the directories you've created, but you can't. When opening the directory info in Finder, I get the old "You have custom access" notice with no ability to make any changes. I'm assuming that this is caused because we're using Windows ACLs on the share, but that's just a wild guess. Changing the write permissions for the group through the terminal works fine, but this is unpractical for the deployment and unreasonable to expect from anyone to do. This is the complete smb.conf: [global] encrypt passwords = yes dns proxy = no strict locking = no read raw = yes write raw = yes oplocks = yes max xmit = 65535 deadtime = 15 display charset = LOCALE max log size = 10 syslog only = yes syslog = 1 load printers = no printing = bsd printcap name = /dev/null disable spoolss = yes smb passwd file = /var/etc/private/smbpasswd private dir = /var/etc/private getwd cache = yes guest account = nobody map to guest = Bad Password obey pam restrictions = Yes # NOTE: read smb.conf. directory name cache size = 0 max protocol = SMB2 netbios name = freenas workgroup = COMPANY server string = FreeNAS Server store dos attributes = yes hostname lookups = yes security = user passdb backend = ldapsam:ldap://ldap.company.local ldap admin dn = cn=admin,dc=company,dc=local ldap suffix = dc=company,dc=local ldap user suffix = ou=Users ldap group suffix = ou=Groups ldap machine suffix = ou=Computers ldap ssl = off ldap replication sleep = 1000 ldap passwd sync = yes #ldap debug level = 1 #ldap debug threshold = 1 ldapsam:trusted = yes idmap uid = 10000-39999 idmap gid = 10000-39999 create mask = 0666 directory mask = 0777 client ntlmv2 auth = yes dos charset = CP437 unix charset = UTF-8 log level = 1 [share] path = /mnt/zfs0 printable = no veto files = /.snap/.windows/.zfs/ writeable = yes browseable = yes inherit owner = no inherit permissions = no vfs objects = zfsacl guest ok = no inherit acls = Yes map archive = No map readonly = no nfs4:mode = special nfs4:acedup = merge nfs4:chown = yes hide dot files force directory mode = 0775 force create mode = 0660

  Read the article

• Transient network dropout for Xen DomU's

  - by Stephen C

  We've got a CentOS server running a cluster of virtuals. Occasionally the cluster's internal network drops out for a minute or so ... and then comes back. The problem is somehow related to the actual network traffic, but it is not a simple load issue. (The system is generally lightly loaded, and the problem occurs irrespective of actual load.) The setup: CentOS 5.6 on Dom0, various CentOS on the DomU's Hardware - a Dell R710 with a BroadCom NextXpress 2 NIC (sigh) using the latest drivers for the NIC from BroadCom Xen configured to use network-bridge and vif-bridge Some iptable tweaks to route an unrelated port to one of the virtuals. The system has one externally visible IP address, and Dom0 runs an Apache httpd configured with a number of virtual hosts each of which reverse proxies to web servers running on the virtuals. (The virtuals have to be NAT'ed, primarily because we don't have enough allocated public IP addresses.) The symptoms: Works fine most of the time. When someone tries to UPLOAD a large file to one virtuals, the internal network drops out ... for all virtuals: The Dom0 httpd sees a network timeout talking to the backend server on the virtual and reports a 502. A previously established ssh connection from Dom0 to any of the DomU's freezes. Our monitoring shows ping failures for traffic between virtuals. The Xen consoles to the DomU's do not freeze. No log messages in any log files that I can see, on either Dom0 or the DomU's ... apart from the Dom0 httpd logs. After a minute or so, the problem clears by itself. This is 100% reproducible. What we've tried: Downloading, building and installing the latest BNX2 driver on Dom0 Turning off MSI on the NIC - adding "options bnx2 disable_msi=1" to /etc/modprobe.conf Turning off tcp segmentation offload - "ethtool -K eth0 tso off". Sacrificing a black rooster at midnight. I've exhausted all my options apart from switching to KVM ... or slaughtering more roosters. Any suggestions?

  Read the article

• Monitoring tools that can take high rate and high volume?

  - by Jon Watte

  We're using Cacti with RRDTool to monitor and graph about 100,000 counters spread across about 1,000 Linux-based nodes. However, our current setup generally only gives us 5-minute graphs (with some data being minute-based); we often make changes where seeing feedback in "near real time" would be of value. I'd like approximately a week of 5- or 10-second data, a year of 1-minute data, and 5 years of 10-minute data. I have SSD disks and a dual-hexa-core server to spare. I tried setting up a Graphite/carbon/whisper server, and had about 15 nodes pipe to it, but it only has "average" for the retention function when promoting to older buckets. This is almost useless -- I'd like min, max, average, standard deviation, and perhaps "total sum" and "number of samples" or perhaps "95th percentile" available. The developer claims there's a new back-end "in beta" that allows you to write your own function, but this appears to still only do 1:1 retention (when saving older data, you really want the statistics calculated into many streams from a single input. Also, "in beta" seems a little risky for this installation. If I'm wrong about this assumption, I'd be happy to be shown my error! I've heard Zabbix recommended, but it puts data into MySQL or some other SQL database. 100,000 counters on a 5 second interval means 20,000 tps, and while I have an SSD, I don't have an 8-way RAID-6 with battery backup cache, which I think I'd need for that to work out :-) Again, if that's actually something that's not a problem, I'd be happy to be shown the error of my ways. Also, can Zabbix do the single data stream - promote with statistics thing? Finally, Munin claims to have a new 2.0 coming out "in beta" right now, and it boasts custom retention plans. However, again, it's that "in beta" part -- has anyone used that for real, and at scale? How did it perform, if so? I'm almost thinking about using a graphing front-end (such as Graphite) and rolling my own retention backend with a simple layer on top of mmap() and some stats. That wouldn't be particularly hard, and would probably perform very well, letting the kernel figure out the balance between frequency of flushing to disk and process operations. Any other suggestions I should look into? Note: it has to have shown itself able to sustain the kinds of data loads I'm suggesting above; if you can point at the specific implementation you're referencing, so much the better!

  Read the article

• nginx + Jetty - thousands of connections stuck in LAST_ACK

  - by virulence

  I have a FreeBSD machine with jails -- two in particular, one that runs nginx and another that runs a Java program that accepts requests via Jetty (embedded mode) Jetty receives upwards of 500 requests/sec constantly and there has been an issue lately where I will constantly have over 60,000 connections in the LAST_ACK state between nginx and jetty. Distribution of all connections (includes some other services, particularly php-fpm) root@host:/root # netstat -an > conns.txt root@host:/root # cat conns.txt | awk '{print $6}' | sort | uniq -c | sort -n 18 LISTEN 112 CLOSING 485 ESTABLISHED 650 FIN_WAIT_2 1425 FIN_WAIT_1 3301 TIME_WAIT 64215 LAST_ACK Distribution of nginx - jetty connections root@host:/root # cat conns.txt | grep '10.10.1.57' | awk '{print $6}' | sort | uniq -c | sort -n 1 3 CLOSE_WAIT 3 LISTEN 18 FIN_WAIT_2 125 ESTABLISHED 64193 LAST_ACK I'd prefer every request to fully close the connection. Clients requests are about 10 minutes apart from each other so connections must be closed. Some of the connections, tcp4 0 0 10.10.1.50.46809 10.10.1.57.9050 LAST_ACK tcp4 0 0 10.10.1.50.46805 10.10.1.57.9050 LAST_ACK tcp4 0 0 10.10.1.50.46797 10.10.1.57.9050 LAST_ACK tcp4 0 0 10.10.1.50.46794 10.10.1.57.9050 LAST_ACK tcp4 0 0 10.10.1.50.46790 10.10.1.57.9050 LAST_ACK tcp4 0 0 10.10.1.50.46789 10.10.1.57.9050 LAST_ACK tcp4 0 0 10.10.1.50.46771 10.10.1.57.9050 LAST_ACK etc.. On Jetty's end I've set maxIdleTime to 2000 -- before this all connections were in ESTABLISHED but they are now LAST_ACK On Jetty's end I've set Connection: close (i.e response.setHeader(HttpHeaders.CONNECTION, HttpHeaderValues.CLOSE);) Jetty never reports a lot of open connections -- always very few. PF/IPFW is not currently being used nginx - reset_timedout_connection is on I cannot figure out how to get nginx or jetty to forcibly close the connection, is this simply something that needs to be fixed in Jetty so that it fully closes the socket after the request finishes? Thanks a lot in advance EDIT: forgot my nginx config for the proxy setup- proxy_pass http://10.10.1.57:9050; proxy_set_header HTTP_X_GEOIP $http_x_geoip; proxy_set_header GEOIP_COUNTRY_CODE $geoip_country_code; proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header Host $http_host; proxy_set_header Connection ""; proxy_http_version 1.1; EDIT2: Forcing Jetty to close the connection via request.getConnection().getEndPoint().close() does nothing -- it's obvious the connection IS being closed (as it's in LAST_ACK) but why isn't it getting past this? Is Nginx keeping the connection open to the backend for some reason?

  Read the article

• Is this distributed database server idea feasible?

  - by David

  I often use SQLite for creating simple programs in companies. The database is placed on a file server. This works fine as long as there are not more than about 50 users working towards the database concurrently (though depending on whether it is reads or writes). Once there are more than this, they will notice a slowdown if there are a lot of concurrent writing on the server as lots of time is spent on locks, and there is nothing like a cache as there is no database server. The advantage of not needing a database server is that the time to set up something like a company Wiki or similar can be reduced from several months to just days. It often takes several months because some IT-department needs to order the server and it needs to conform with the company policies and security rules and it needs to be placed on the outsourced server hosting facility, which screws up and places it in the wrong localtion etc. etc. Therefore, I thought of an idea to create a distributed database server. The process would be as follows: A user on a company computer edits something on a Wiki page (which uses this database as its backend), to do this he reads a file on the local harddisk stating the ip-address of the last desktop computer to be a database server. He then tries to contact this computer directly via TCP/IP. If it does not answer, then he will read a file on the file server stating the ip-address of the last desktop computer to be a database server. If this server does not answer either, his own desktop computer will become the database server and register its ip-address in the same file. The SQL update statement can then be executed, and other desktop computers can connect to his directly. The point with this architecture is that, the higher load, the better it will function, as each desktop computer will always know the ip-address of the database server. Also, using this setup, I believe that a database placed on a fileserver could serve hundreds of desktop computers instead of the current 50 or so. I also do not believe that the load on the single desktop computer, which has become database server will ever be noticable, as there will be no hard disk operations on this desktop, only on the file server. Is this idea feasible? Does it already exist? What kind of database could support such an architecture?

  Read the article

• Squid on windows loadbalancing only to one server

  - by Martin L.

  After thousands of googles and trying days i cant get the load balancer/failover in squid on windows to work. Iam using squid 2.7. My webservers are 2 single NIC lighttpd and one dual nic lighttpd. server1 in this example is running squid on port 80 and lighttpd on port 8080 (just to test) Requirements: All 3 webservers running lighttpd should be balanced two option for load balancing: Best would be if server1 is busy server2 takes over, if server2 is busy server3 takes over, etc.. Round robin style evenly distributed load. Eg server1 takes first call, server2 second etc.. All requests should be treated the same way (no url rewriting or so on) Sent host headers have to be redirected to every server as http host header, speaking of "server1", "server1.company.internal" and "10.211.1.1". My approach: acl all src all acl manager proto cache_object http_port 80 accel defaultsite=server1.company.internal vhost #reverse proxy entries cache_peer 10.211.2.1 parent 8080 0 no-query originserver round-robin login=PASS name=server1_nic1 cache_peer 10.211.1.2 parent 80 0 no-query originserver round-robin login=PASS name=server2_nic1 cache_peer 10.211.2.3 parent 8080 0 no-query originserver round-robin login=PASS name=server3_nic1 cache_peer 10.211.2.4 parent 8080 0 no-query originserver round-robin login=PASS name=server3_nic2 #decl of names of squid host acl registered_name_hostdomain dstdomain server1.company.internal acl registered_name_host dstdomain server1 #ip of squid host acl registered_name_ip dstdomain 10.211.2.1 # access: redirects the correct squid hostname http_access allow registered_name_hostdomain http_access allow registered_name_host http_access allow registered_name_ip http_access deny all cache_peer_access server1_nic1 allow registered_name_hostdomain cache_peer_access server1_nic1 allow registered_name_host cache_peer_access server1_nic1 allow registered_name_ip cache_peer_access server2_nic1 allow registered_name_hostdomain cache_peer_access server2_nic1 allow registered_name_host cache_peer_access server2_nic1 allow registered_name_ip cache_peer_access server3_nic1 allow registered_name_hostdomain cache_peer_access server3_nic1 allow registered_name_host cache_peer_access server3_nic1 allow registered_name_ip cache_peer_access server3_nic2 allow registered_name_hostdomain cache_peer_access server3_nic2 allow registered_name_host cache_peer_access server3_nic2 allow registered_name_ip cache_peer_access server1_nic1 deny all cache_peer_access server2_nic1 deny all cache_peer_access server3_nic1 deny all cache_peer_access server3_nic2 deny all never_direct allow all Problems: Load balancer does not load balance other than to first server. Only if the first server is killed in any way the second will take over. I have seen the others working at some point, but definitely not as the intended load balancing described above. If the cache_peer_access is not defined sometimes the wrong hostname is sent to the backend webserver and this always depends on the defaultsite= parameter. Probably because the host header on the request to squid is not set and its replaced by defaultsite. Leaving out defaultsite didnt solve the problem. The only workaround i found for this is the current approach with cache_peer_access. Questions: Does the cache_peer_access influence the round-robin? Is there a better workaround to pass the host header to the backed webservers? Which parameters do increase the speed of load balancing or does anyone have a better approach? -Martin

  Read the article

• Varnish server in front of nginx server with multiple virtualhosts

  - by Garreth 00

  I have tried to search for a solution for this, but can't find any documentation/tips on my specific setup. My setup: Backendserver: ngnix: 2 different websites (2 top domains) in virtualenv, running gunicorn/python/django Backendserver hardware(VPS) 2gb ram, 8 CPU Databaseserver: postgresql - pg_bouncer Backendserver hardware (VPS) 1gb ram, 8 CPU Varnishserver: only running varnish Varnishserver hardware (VPS) 1gb ram, 8 CPU I'm trying to set up a varnish server to handle rare spike in traffic (20 000 unique req/s) The spike happens when a tv program mention one of the sites. What do I need to do, to make the varnish server cache both sites/domains on my backendserver? My /etc/varnish/default.vcl : backend django_backend { .host = "local.backendserver.com"; .port = "8080"; } My /usr/local/nginx/site-avaible/domain1.com upstream gunicorn_domain1 { server unix:/home/<USER>/.virtualenvs/<DOMAIN1>/<APP1>/run/gunicorn.sock fail_timeout=0; } server { listen 80; listen 8080; server_name domain1.com; rewrite ^ http://www.domains.com$request_uri? permanent; } server { listen 80 default_server; listen 8080; client_max_body_size 4G; server_name www.domain1.com; keepalive_timeout 5; # path for static files root /home/<USER>/<APP>-media/; location / { proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header Host $http_host; proxy_redirect off; if (!-f $request_filename) { proxy_pass http://gunicorn_domain1; break; } } } My /usr/local/nginx/site-avaible/domain2.com upstream gunicorn_domain2 { server unix:/home/<USER>/.virtualenvs/<DOMAIN2>/<APP2>/run/gunicorn.sock fail_timeout=0; } server { listen 80; listen 8080; server_name domain2.com; rewrite ^ http://www.domains.com$request_uri? permanent; } server { listen 80; listen 8080; client_max_body_size 4G; server_name www.domain2.com; keepalive_timeout 5; # path for static files root /home/<USER>/<APP>-media/; location / { proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; proxy_set_header Host $http_host; proxy_redirect off; if (!-f $request_filename) { proxy_pass http://gunicorn_domain2; break; } } } Right now, If I try the Ip of the varnishserver I only get served domain1.com. Would everything be correct if I change the DNS of the two domain to point to the varnishserver, or is there extra setup before it would work? Question 2: Do I need a dedicated server for varnish, or could I just install varnish on my backendserver, or would the server run out of memory quick?

  Read the article

• Apache's htcacheclean doesn't scale: How to tame a huge Apache disk_cache?

  - by flight

  We have an Apache setup with a huge disk_cache (500.000 entries, 50 GB disk space used). The cache grows by 16 GB every day. My problem is that the cache seems to be growing nearly as fast as it's possible to remove files and directories from the cache filesystem! The cache partition is an ext3 filesystem (100GB, "-t news") on an iSCSI storage. The Apache server (which acts as a caching proxy) is a VM. The disk_cache is configured with CacheDirLevels=2 and CacheDirLength=1, and includes variants. A typical file path is "/htcache/B/x/i_iGfmmHhxJRheg8NHcQ.header.vary/A/W/oGX3MAV3q0bWl30YmA_A.header". When I try to call htcacheclean to tame the cache (non-daemon mode, "htcacheclean-t -p/htcache -l15G"), IOwait is going through the roof for several hours. Without any visible action. Only after hours, htcacheclean starts to delete files from the cache partition, which takes a couple more hours. (A similar problem was brought up in the Apache mailing list in 2009, without a solution: http://www.mail-archive.com/[email protected]/msg42683.html) The high IOwait leads to problems with the stability of the web server (the bridge to the Tomcat backend server sometimes stalls). I came up with my own prune script, which removes files and directories from random subdirectories of the cache. Only to find that the deletion rate of the script is just slightly higher than the cache growth rate. The script takes ~10 seconds to read the a subdirectory (e.g. /htcache/B/x) and frees some 5 MB of disk space. In this 10 seconds, the cache has grown by another 2 MB. As with htcacheclean, IOwait goes up to 25% when running the prune script continuously. Any idea? Is this a problem specific to the (rather slow) iSCSI storage? Should I choose a different file system for a huge disk_cache? ext2? ext4? Are there any kernel parameter optimizations for this kind of scenario? (I already tried the deadline scheduler and a smaller read_ahead_kb, without effect).

  Read the article

• Nginx Load Balancer 403 error

  - by user64473

  I am trying to install nginx as a load balancer with apache backends, so that when I point my sites to the nginx server it serves up the content from the apache backend. I have the apache configuration set up correctly on both (i.e when I go to the site on the apache servers it works great) but when I use the nginx load balancer as the site I get 403 error. I have no idea why as it isn't even accessing any files on the server, thusly there aren't any files to be forbidden access to. My virtual host is enabled and looks like this: upstream webs { server 10.0.0.30 weight=1; server 10.0.0.31 weight=1; } server { listen 80; server_name www.example.com example.com; access_log /var/log/nginx/access.log; location / { proxy_pass http://webs; include /etc/nginx/proxy.conf; } } and my nginx.conf looks like this: user www-data; worker_processes 4; error_log /var/log/nginx/error.log; pid /var/run/nginx.pid; events { worker_connections 1024; # multi_accept on; } http { include /etc/nginx/mime.types; access_log /var/log/nginx/access.log; sendfile on; #tcp_nopush on; #keepalive_timeout 0; keepalive_timeout 65; tcp_nodelay on; gzip on; gzip_disable "MSIE [1-6]\.(?!.*SV1)"; include /etc/nginx/conf.d/*.conf; include /etc/nginx/sites-enabled/*; proxy_redirect off; proxy_set_header Host $host; proxy_set_header X-Real-IP $remote_addr; proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; client_max_body_size 10m; client_body_buffer_size 128k; proxy_connect_timeout 90; proxy_send_timeout 90; proxy_read_timeout 90; proxy_buffers 32 4k; } Can any geniuses out there tell me what I am doing wrong?

  Read the article

• Surface development: it&rsquo;s just like software development

  - by Dennis Vroegop

  Surface is magic. Everyone using it seems to think that way. And I have to be honest, after working for almost 2 years with the platform I still get that special feeling the moment I turn on the unit to do some more work. The whole user experience, the rich environment of the SDK, the touch, even the look and feel of the Surface environment is so much different from the stuff I’ve been working on all my career that I am still bewildered by it. But… and this is a big but.. in the end we’re still talking about a computer and that needs software to become useful. Deep down the magic of the Surface unit there is a PC somewhere, running Windows Vista and the .net framework 3.5. When you write that magic software that makes the platform come alive you’re still working with .net, WPF/XNA, C#, VB.Net and all those other tools and technologies you know so well. Sure, the whole user experience is different from what you’ve known. And the way of thinking about users, their interaction and the positioning of screen elements requires a whole new paradigm. And that takes time. It took me about half a year before I had the feeling I got it nailed down. But when that moment came (about 18 months ago…) I realized that everything I had learned so far on software development still is true when it comes to Surface. The last 6 months I have been working with some people with a different background to start a new company. The idea was that the new company would be focussing on Surface and Surface only. These people come from a marketing background and had some good ideas for some applications. And I have to admit: their ideas were good. Very good. Where it all fell down of course is that these ideas need to be implemented in a piece of software. And creating great software takes skilled developers and a lot of time and money. That’s where things went wrong: the marketing guys didn’t realize and didn’t want to realize that software development is a job that takes skill. You can’t just hire a bunch of developers and expect them to deliver the best sort of software, especially not when it comes to Surface. I tried to explain that yes, their User Interface in Photoshop looked great, but no: I couldn’t develop an application like that in a weeks time. Even worse: the while backend of the software (WCF for communications, SQL Server for the database, etc) would take a lot more time than the frontend. They didn’t understand. It took them a couple of days to drawn the UI in Photoshop so in Blend I should be able to build the software in about the same amount of time. Well, you and I know that it doesn’t work that way. Software is hard to write, and even harder to write well, and it takes skill and dedication. It’s not something you can do as fast as you can draw a mock up for a Surface application in Photohop. The same holds true for web applications of course. A lot of designers there fail to appreciate the hard work that goes into writing the plumbing for a good web app that can handle thousands of users. Although the UI is very important, it’s not all there is to it. And in Surface development this is the same. The UI should create the feeling of magic, but the software behind it is what makes it come alive. And that takes time. A lot of time. So brush of you skills and don’t throw them away if you start developing for Surface. Because projects (and colaborations) can fail there as hard as they can in any other area of software development. On a side note: we decided to stop the colaboration (something the other parties involved didn’t appreciate and were very angry about) and decided to hire a designer for the Surface projects. The focus is back where it belongs: on the software development we know so well and have been doing very well for 13 years. UI is just a part of the whole project and not the end product. So my company Detrio is still going strong when it comes to develivering Surface solutions but once again from a technological background, not a marketing background.

  Read the article

• Trying to keep up with Technology and Blogging

  - by Dave Campbell

  A little bit of everything... The heading above got changed a bunch during writing and I finally settled on that because this has become a 'stream of consciousness' post... or maybe a stream of UNconsciousness :) If you've noticed, my blogging has been a tad slow this fall. There's been a lot going on personally. But then again, I haven't skipped anybody either. Rather than go through ALL the blogs I have aggregated, and take a week to get to the bottom, at some point in the last year, I had moved the lists around so I now have "SilverlightMVPs", "Very Prolific", "WP7", and "Top Checks". This is a total of about 250 of the more prolific bloggers. Those 250 bloggers have kept me very busy up through about //BUILD. Sometimes it would take all week to go through just that list putting out 13 posts per blog per day... but not anymore. This weekend I made it all the way through the BIG list... close to 700 blogs, and if you read my blog, you know I had one medium day (Saturday), and yesterday was very short. Why is this? To be honest, I don't know... is everybody busy re-tooling, or churning waiting for direction? I have a short list of WinRT/Metro/W8 folks... maybe I need to be pointed to more of them... but my old favorites are not pumping out posts as they have in the past. I said before that I am attracted to Metro, and I've already got My first Metro app post out there, and were it not for working with the new site, I'd have had another out last weekend... so definitely look for more from me in that area. New Site? Did I say 'new site' ? oops... didn't mean to do that, but now that the cat is out of the bag, I may as well continue... While at //BUILD, I discussed a re-tooling of SilverlightCream with lots of folks... probably more than wanted to hear about it to be honest! ... it's needed a facelift, and there's stuff on there that never worked right, plus there's a lot of manual effort that goes into a blog post. In an effort to alleviate all the above, Michael Washington and I have been working on the next iteration of SilverlightCream. Not wanting to lose that branding or mess with any saved links, I decided to change from a somewhat funky name to something more professional. I also decided to put my blog on the site, and tie my main announcement twitter feed to the site as well. The way things sit today, there are 3 different names in those locations and it's gotta be confusing for folks just stumbling in. We're going to do a series of posts talking about the site and the new backend processing (hint: Michael Washington is responsible for it, so you can take a guess at the technology), but for now, we'd like some eyes on the front end of the site, and some submittals using it to see if it falls over somewhere that we haven't tried. So... I'm going to give it up... the new site is Windows Dev News. The Twitter feed is @WindowsDevNews, and the blog will be on the site as well at Windows Dev News Blog. I've got the RSS Feed on Feedburner too, so I think all the nuts and bolts are good to go. The submittal and search pages work, as does the blog page. You'll notice we used the MasterPage from SilverlightCream to get started. That will probably change, but it's just the visual... the content is the important part. Other missing things are the tracking and 'Skim' page that we will eventually have up and running. There are some formatting issues with the blog posts but if you hang in there with me, those will be taken care of. If you're a blogger, please submit through the site and let me know if you find any problems. If you're a reader, please add this feed and site. I'll be duplicating the effort for a while but at some point will stop that foolishness. We won't lose the data from SilverlightCream though, so keep using that as a search resource... I have hopes to pull that database over to WindowsDevNews, or link to it in some manner... that part isn't set in jello yet, but it will not be lost. So there it is... let me know what you think, send me your WinRT/Metro/W8 postings along with your Silverlight and WP7 posts... it's not that different, it's just more. Stay in the 'Light

  Read the article

< Previous Page | 64 65 66 67 68 69 70 71 72 73 74 75  | Next Page >