Search Results

Search found 13075 results on 523 pages for 'bounding box'.

Page 404/523 | < Previous Page | 400 401 402 403 404 405 406 407 408 409 410 411  | Next Page >

• Password-protected sharing allows access to users who have no account?

  - by romkyns

  Running Win7 on two computers in my LAN. Computer A has password-protected sharing enabled, and shares a folder. It has a single user account "Bob", and the Guest account is turned off. The network is workgroup-based. According to the descriptions of the "password-protected sharing" I could find, the only people who can access the shared folder via the LAN are those who know the username+password for the "Bob" account. However a second computer on the LAN is able to view this shared folder by simply browsing to Computer A. They don't need to enter any passwords or anything. The only user account registered on that PC is called "Jim", and has a different password from "Bob". How on earth is computer B able to view this shared folder? Is the popular description of the "password-protected sharing" feature inaccurate / did I misunderstand it big time? P.S. There is a possibility that the password for "Bob" has been entered on that PC once, and possibly the "remember password" box was checked. I've looked in the "Credential Manager" on both computers and there is nothing saved anywhere.

  Read the article

• Determining a realistic measure of requests per second for a web server

  - by Don

  I'm setting up a nginx stack and optimizing the configuration before going live. Running ab to stress test the machine, I was disappointed to see things topping out at 150 requests per second with a significant number of requests taking 1 second to return. Oddly, the machine itself wasn't even breathing hard. I finally thought to ping the box and saw ping times around 100-125 ms. (The machine, to my surprise, is across the country). So, it seems like network latency is dominating my testing. Running the same tests from a machine on the same network as the server (ping times < 1ms) and I see 5000 requests per second, which is more in-line with what I expected from the machine. But this got me thinking: How do I determine and report a "realistic" measure of requests per second for a web server? You always see claims about performance, but shouldn't network latency be taken into consideration? Sure I can serve 5000 request per second to a machine next to the server, but not to a machine across the country. If I have a lot of slow connections, they will eventually impact my server's performance, right? Or am I thinking about this all wrong? Forgive me if this is network engineering 101 stuff. I'm a developer by trade. Update: Edited for clarity.

  Read the article

• SQL Server Backup problem when browsing to the directory

  - by Richard West

  I want to allow a group (eg. 'BackupManagers') who can only preform backup and restore operations on certain databases. When creating the BackupManagers user account I checked db_backupoperator. When the user logs in to create a backup they get an error message similar to the following when the select Tasks - Backup - Click on Add in the destiantion block - click on the "..." button to browse TITLE: Locate Database Files - MYSERVER\SQL2005 E:\MSSQL\Backup Cannot access the specified path or file on the server. Verify that you have the necessary security privileges and that the path or file exists. If you know that the service account can access a specific file, type in the full path for the file in the File Name control in the Locate dialog box. I have confirmed that the user has permissions to the folder. I have even created a share to this folder and had them access it through explorer. They are able to create and delete files within the folder. I have found that if they type in the path to the file instead of using the "..." button to browse the directory tree then they can create a backup file fine. Why is the browse button not working as expected? Thanks!

  Read the article

• Why do I get a DegradedArray event with mdadm

  - by azera

  Hello Just so we're clear on what's happening: I bought 4 new sata 2 drives, with the intent of using them in a raid5 all drive are fully recognised by both my bios and my linux box (gentoo) I created a raid5 array, fiddled a bit with it to understand how it works, how to monitor ect At some point, this triggered a degradedarray event, even though the array is brand new. I tried to stopping the array and recreating a new array with the same drive but the new array starts degraded too. here is what I used to create it mdadm --create -l5 -n4 /dev/md/md0-r5 /dev/sdb /dev/sdd /dev/sde /dev/sdf here are the output from my /proc/mdstat and mdadm --detail --scan **mdstat** Personalities : [raid0] [raid1] [raid6] [raid5] [raid4] [raid10] md127 : active raid5 sdf[4] sde[2] sdd[1] sdb[0] 4395415488 blocks level 5, 64k chunk, algorithm 2 [4/3] [UUU_] [>....................] recovery = 2.8% (41689732/1465138496) finish=890.3min speed=26645K/sec unused devices: <none> **detail** ARRAY /dev/md/md0-r5 metadata=0.90 spares=1 UUID=453e2833:81f22a74:64188b84:66721085 As such I have a couple questions: does a raid5 array always start in degraded mode at first ? why does sdf have the number 4 between bracket instead of 3, why does it see a spare disk and why is the 4th drive marked with _ instead of U ? (bad configuration ?) How can I recreate the array from scratch, do i have to format each drive on its own before recreating it ? Thanks for any help, I'm not sure about what I should do at the moment

  Read the article

• Virtualbox - differencing disk based on different differencing disk

  - by Klinki

  I'm trying to create differencing image based on differencing image in VirtualBox 4.2.18. Official documentation says it should be possible: http://www.virtualbox.org/manual/ch05.html#diffimages Basically I want to achieve this drive hierarchy: + immutable image with Debian and all software installed +---- differencing image with specific configuration, autoreset=off, readonly +-------- differencing image with autoreset=on +---- another differencing image for different virtual machine +-------- differencing image with autoreset=on I successfully created differencing image based on differencing image, but I'm not able to connect it to virtual machine :( It always shows error: Failed to open the hard disk .... cannot register hard disk ... because hard disk with UUID ... already exists Here is screenshot of Virtual Media Manager and error dialog Virtual Media Manager Window screenshot Very strange is that the new differencing image (tempdrive.vdi) doesn't have Actual Size 0. I wasn't able to connect it, but still, it has 36KB of data on it... This is very similar to this older question: How to create a chained differencing disk of another differencing disk in Virtual Box? but suggested solution is not working anymore in VirtualBox 4.2.18, so I posted it as a new question. (Limit for posting links and screenshots is quite annoying..)

  Read the article

• how to automatically mount ~/Private using ecryptfs when logging in via ssh pubkey

  - by andreash

  Raionale: I want to be able to automatically make backups to a remote machine, which will be encrypted with ecryptfs. The title says it all: I set up ecryptfs-utils on my Debian Squeeze box, and set up one user to use it via ecryptfs-setup-private. When I log in via SSH using password authentication, the ~/Private directory automatically gets mounted. How can I achieve that ~/Private also automatically gets mounted when logging in via SSH using public key authentication? Obviously, the best solution would be if ecryptfs could somehow 'use' the SSH public key to en/decrypt the data (I know that then using the user's password would not be able to en/decrypt the data any more; this would be acceptable). Probably, this will not work. So perhaps somehow call ecryptfs-mount-private via ssh before logging in via public key? Probably, then I would need to somehow pipe the passphrase through the SSH connection, right? So I would need to store it on the source machine's file system. Not nice either. Any other ideas?

  Read the article

• Rebuild Apple RAID set

  - by Clinton Blackmore

  We have a Mac Pro tower with an Apple RAID card in it using third party drives. When one drive failed, we replaced it and the RAID 5 set was nearly done rebuilding when the computer was rebooted. It did not come back up. We are now booting up off of a different internal volume, and have three (third-party) drives of identical spec (including revision and firmware) in the box. One of the drives is a global spare; the other two are recognized as belong to a RAID set but are in "Roaming" mode. The intention is to recreate the three-drive RAID set using the data on the two drives that are good. When we tell the system to create a RAID 5 using the three drives, it tells us that it'll create a RAID set but everything will be lost. There are no obvious options to rebuild a RAID using the two good drives and incorporating the third drive in Apple's RAID Utility, and we've looked through the options for the raidutil command. Fortunately, all important data is backed up, and we can rebuild from scratch, but, is there any way to make the RAIDset work again?

  Read the article

• Name one good reason for immediately failing on a SMTP 4xx code

  - by Avery Payne

  I'm really curious about this. The question (highlighed in bold): Can someone name ONE GOOD REASON to have their email server permanently set up to auto-fail/immediate-fail on 4xx codes? Because frankly, it sounds like "their" setups are broken out-of-the-box. SMTP is not Instant Messaging. Stop treating it like IRC or Jabber or MSN or insert-IM-technology-here. I don't know what possesses people to have the "IMMEDIATE DELIVERY OR FAIL" mentality with SMTP setups, but they need to stop doing that. It just plain breaks things. Every two or three years, I stumble into this. Someone, somewhere, has decided in their infinite wisdom that 4xx codes are immediate failures, and suddenly its OMGWTFBBQ THE INTARNETZ ARE BORKEN, HALP SKY IS FALLING instead of "oh, it'll re-attempt delivery in about 30 minutes". It amazes me how it suddenly becomes "my" problem that a message won't go through, because someone else misconfigured "their" SMTP service. IF there is a legitimate reason for having your server permanently set up in this manner, then the first good answer will get the check. IF there is no good reason (and I suspect there isn't), then the first good-sounding-if-still-logically-flawed answer will get the check.

  Read the article

• Static noise in headphones

  - by John Murdoch

  I have a Asus P6T based system. I was using the on-board sound (plugging in Logitech X-230 2.1 analog speakers in the green "front speakers" 3.5mm analog output, then plugging in my headphones in that). I was quite happy with the sound quality (didn't hear any static noise if volume was turned down to my normal listening level). Then about a week ago I started having terrible static noise from the left channel, and no normal audio on that left channel. Right channel had more static noise than usual but did have a bit of sound. I tried using the AC'97 in front of my case but that seemed to have no signal. I decided my on-board sound card has gone bad and bought an internal sound card to replace it (Startech 7.1Ch PCI). This fixed the "no sound from left channel problem", but I had much more audible static noise. I decided the card was low quality and/or it had interference from all the other things happening inside the computer case, and bought a Sweex SC016 external USB sound card. But even with that I have static noise in headphones. Positioning the USB sound card differently doesn't seem to help. Trying the other analog outputs (e.g., surround) doesn't help. The static noise in all cases is proportional to the volume. I have tried different headphones, but the situation is situation though perhaps the flavour of the static noise changes slightly. So what are my options? a) Get another, more expensive, external USB sound card hoping the quality will improve? b) Get another, more expensive, internal sound card (PCIe 1x perhaps) hoping the quality will improve? c) Get a dedicated DAC box? d) Get some Hi-Fi earphones? Suggestions? tl;dr - Two different sound cards both still have static noise in headphones.

  Read the article

• How to send Content-Disposition headers in apache for files?

  - by Rory McCann

  I have a directory of text files that I'm serving out with apache 2. Normally when I (or any user) access the files they see them in their browser. I want to 'force'* the web browser to pop up a 'Save as' dialog box. I know this is possible to do with the Content-Disposition headers (more info). Is there some way to turn that on for each file? Ideally I'd like something like this: <Directory textfiles> AutoAddContentDispositionHeaders On </Directory> And then apache would set the correct content disposition header, including using the same filename. Something like this might be possible with the apache Header directive. Bonus points if it's included by standing in apache in debian. I could do a simple PHP wrapper script that takes in a filename argument, makes the call to header(...) and then prints the file, but then i have to validdate input etc. that's work I'm trying to avoid. * I know you can't actually force things when it comes to the web

  Read the article

• Xvnc4 started from xinetd only displays empty gray X screen

  - by scott8035

  I'm attempting to setup an Ubuntu 10.10 box so that anyone can connect to port 5900 and be greeted by the gdm login manager. To do so, I added a vnc entry in /etc/services and I am starting Xvnc4 using this xinetd config file: service vnc { protocol = tcp socket_type = stream wait = no user = nobody server = /usr/bin/Xvnc server_args = -geometry 1000x700 -depth 24 -broadcast -inetd -once -securitytypes None } This kind of works...I can start multiple sessions all to port 5900, and I get an X screen. The problem is that I only get an empty, gray X screen with no applications started. I know when you run vncserver from the command line it will look to your ~/.vnc/ directory for your passwd and xstartup files, and I think what I want to do is put "gnome-session" into the xstart file. However, which xstartup file? The running user is "nobody" who obviously doesn't have a ~/.vnc/ directory. I tried a /root/.vnc/xstartup file and a ~scott/.vnc/xstartup file and it doesn't look like they were even read. I changed the xinetd vnc service so that it would "strace" Xvnc4. I looked thru all the "open" lines and didn't get a clue as to what file it was trying to read for xstart. Can anyone help? I just want a terminal server where the user is presented with a gdm login screen.

  Read the article

• Routing for Two Hosts Behind a IPSec Tunnel

  - by Brent

  Network A 10.110.15.0/24 Firewall is .1 Host A is .2 Network B 10.110.16.0/24 Firewall is .1 Host B is .2 Two Cisco ASA's. IPSec tunnel with a crypo map that secures 10.110.15.0/24 <- 10.110.16.0/24. Let's say two hosts, 10.110.15.2 and 10.110.16.2 need to talk to each other. Normally I have to enter a persistent static route on a each host along the lines of: route add 10.110.16.0 mask 255.255.255.0 10.110.15.1 metric 1 -p (on the "A" box) I also have to enter another persistent static route on the .16 host in order for the traffic to know how to get back to the .15 network. Note that the default for each machine IS the firewall, so .1. I have no problem adding persistent routes on Windows/ESX/*nux machines but what about a smart switch in the .16 network that I want to manage from the .15 network. Do I need to run a routing protocol? Do I need to have Reverse Route Injection enabled on both ends of the IPSec tunnel? Should I add a route on the firewall? If so, how do you formulate it? Does it get a metric of 1 and my default route 0.0.0.0 get a metric of 2?

  Read the article

• Server load high, CPU idle. NFS the cause?

  - by Mech Software

  I am running into a scenario where I'm seeing a high server load (sometimes upwards of 20 or 30) and a very low CPU usage (98% idle). I'm wondering if these wait states are coming as part of an NFS filesystem connection. Here is what I see in VMStat procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu------ r b swpd free buff cache si so bi bo in cs us sy id wa st 2 1 0 1298784 0 0 0 0 16 5 0 9 1 1 97 2 0 0 1 0 1308016 0 0 0 0 0 0 0 3882 4 3 80 13 0 0 1 0 1307960 0 0 0 0 120 0 0 2960 0 0 88 12 0 0 1 0 1295868 0 0 0 0 4 0 0 4235 1 2 84 13 0 6 0 0 1292740 0 0 0 0 0 0 0 5003 1 1 98 0 0 4 0 0 1300860 0 0 0 0 0 120 0 11194 4 3 93 0 0 4 1 0 1304576 0 0 0 0 240 0 0 11259 4 3 88 6 0 3 1 0 1298952 0 0 0 0 0 0 0 9268 7 5 70 19 0 3 1 0 1303740 0 0 0 0 88 8 0 8088 4 3 81 13 0 5 0 0 1304052 0 0 0 0 0 0 0 6348 4 4 93 0 0 0 0 0 1307952 0 0 0 0 0 0 0 7366 5 4 91 0 0 0 0 0 1307744 0 0 0 0 0 0 0 3201 0 0 100 0 0 4 0 0 1294644 0 0 0 0 0 0 0 5514 1 2 97 0 0 3 0 0 1301272 0 0 0 0 0 0 0 11508 4 3 93 0 0 3 0 0 1307788 0 0 0 0 0 0 0 11822 5 3 92 0 0 From what I can tell when the IO goes up the waits go up. Could NFS be the cause here or should I be worried about something else? This is a VPS box on a fiber channel SAN. I'd think the bottleneck wouldn't be the SAN. Comments?

  Read the article

• Uninstall Perl 5.10 and Install 5.8 on Ubuntu 10.04

  - by user48848

  I was using Ubuntu 8.04 with installed Perl 5.8.8. I had installed Bugzilla 3.4.4 with some custom scripts and it was working fine. I had to upgrade Ubuntu from 8.04 to 10.04 and automatically Perl is also upgraded from Perl 5.8 to 5.10. Now, accessing Bugzilla displays an error message related to Perl module. While trying to run checksetup.pl, I am getting the following error message: root@cvsnew:/var/www/bugzilla-3.4# perl checksetup.pl * This is Bugzilla 3.4.3 on perl 5.10.1 * Running on Linux 2.6.32-24-generic-pae #39-Ubuntu SMP Wed Jul 28 07:39:26 UTC 2010 Checking perl modules... Checking for CGI.pm (v3.33) ok: found v3.48 perl: symbol lookup error: lib/i486-linux-gnu-thread-multi/auto/Digest/SHA/SHA.so: undefined symbol: Perl_Tstack_sp_ptr Accessing Bugzilla throws : "500 Internal Server Error". I explored this issue and found that this issue has come up because upgrading Ubuntu upgrades Perl 5.8 to 5.10 automatically. It seems like Perl 5.10.1 isn't binary compatible with the previous. Now, I want to uninstall Perl 5.10.1 and Install Perl 5.8.8 on my Ubuntu 10.04 box. I will appreciate if someone has already done that or share the required steps/commands to do it.

  Read the article

• Dual boot windows 8 pro and windows 7 on XPS 8500 Special Edition

  - by Jesse

  I am trying to install a dual boot with windows 7 premium and windows 8 Pro on an XPS 8500 special edition. I created a new primary partition on my C: drive, inserted the windows 8 install disk, and rebooted my computer from DVD. I select custom install and the dialog box saying "Where do you want to install windows at?" pops up but none of my drives are listed. Please help me determine what is going on. I don't understand why none of my drives are showing up on this menu. Not even the original drive. When I go to load driver and click on the partition I created it tells me "No signed device drivers were found. Make sure the installation media contains the correct drivers, and then click OK." resolved above issue by running setup from the source folder on the install disk instead of booting from DVD. Was able to locate my new partition and start install. It completes the first step of "Copying windows files" just fine but then on the next step "Getting files ready for installation" my computer restarts and attempts to load windows 8 but keeps telling me my pc needs to restart. This keeps going on in an infinite boot loop. Please help, this has been a nightmare!

  Read the article

• Some Can reach bidmail.com others can't.

  - by user69426

  On a windows 7 Professional machine in Chrome one of our Estimating assistants can't get to www.bidmail.com, however the other 3 can. On his machine I did nslookup then bidmail.com and it fails to find it. I then went to a machine that could reach bidmail and did nslookup. It can't find it. I was skeptical and thought maybe it was a cached page so I cleared the cache then went back to bidmail.com was able to get to the page, login, lookup a newly posted bid then download the file. Yet I can not look it up through nslookup and I can't ping it www.bidmail.com and I can't trace it. I remoted to our other warehouse which is set up as a workgroup and attempt to nslookup bidmail and that nslookup fail... and on that machine which has never been to bidmail before it was able to connect to the website! I am totally confused if I can't ping it and I can't use nslookup to get there how in the hell is Chrome getting to the page and how do I get this guy back on? Also while typing this I took a new laptop out of the box plugged it in with no updates and can get to bidmail! omg!

  Read the article

• HTB.init / tc behind NAT

  - by Ben K.

  I have an Ubuntu 10 box that I'm trying to set up as a bandwidth-shaping router. The machine has one WAN interface, eth0 and two LAN interfaces, eth1 and eth2. NAT is configured using MASQUERADE as described at InternetConnectionSharing. I'm mostly concerned with shaping outbound traffic from the LAN interfaces -- in the end, I'd like to end up with a hard 768Kbps limit per-LAN-interface (rather than a limit on eth0 pooled across all interfaces). I installed HTB.init, and riffing on the examples, tried to set this up on eth1 by putting three files into /etc/sysconfig/htb: /etc/sysconfig/htb/eth1 DEFAULT=30 R2Q=100 /etc/sysconfig/htb/eth1-2.root RATE=768Kbps BURST=15k /etc/sysconfig/htb/eth1-2:30.dfl RATE=768Kbps CEIL=788Kbps BURST=15k LEAF=sfq I can /etc/init.d/htb start and /etc/init.d/htb stats and see information that /seems/ to suggest it's working...but when I try pulling a large file via the WAN interface the shaping clearly isn't in effect. Any suggestions? My guess is it has something to do with where the shaping falls in the NAT chain, but I really have no idea where to begin troubleshooting this. ---- Update: Here's my /etc/init.d/htb list output, it seems to make sense -- the default rate for eth1 is 768Kbps? ### eth0: queueing disciplines qdisc htb 1: root refcnt 2 r2q 100 default 30 direct_packets_stat 0 qdisc sfq 30: parent 1:30 limit 127p quantum 1514b perturb 10sec ### eth0: traffic classes class htb 1:2 root rate 768000bit ceil 768000bit burst 1599b cburst 1599b class htb 1:30 parent 1:2 leaf 30: prio 0 rate 6144Kbit ceil 6144Kbit burst 15Kb cburst 1598b ### eth0: filtering rules filter parent 1: protocol ip pref 100 u32 filter parent 1: protocol ip pref 100 u32 fh 800: ht divisor 1 filter parent 1: protocol ip pref 100 u32 fh 800::800 order 2048 key ht 800 bkt 0 flowid 1:30 match 00000000/00000000 at 12 match 00000000/00000000 at 16 ### eth1: queueing disciplines qdisc htb 1: root refcnt 2 r2q 100 default 30 direct_packets_stat 0 qdisc sfq 30: parent 1:30 limit 127p quantum 1514b perturb 10sec ### eth1: traffic classes class htb 1:2 root rate 768000bit ceil 768000bit burst 1599b cburst 1599b class htb 1:30 parent 1:2 leaf 30: prio 0 rate 6144Kbit ceil 6144Kbit burst 15Kb cburst 1598b

  Read the article

• hosts file seems to be ignored

  - by z4y4ts

  I have almost fresh Ubuntu desktop box. OS was installed two weeks ago and updated from karmic repositories. Last week I had no problems with DNS. But this week something had changed. I'm not sure what and when, and not sure whether I changed any configs. So now I have some really weird situation. According to logs name resolving should work normally. /etc/hosts 127.0.0.1 localhost test 127.0.1.1 desktop /etc/host.conf order hosts,bind multi on /etc/resolv.conf # Generated by NetworkManager search search servers obtained via DHCP nameserver 192.168.0.3 /etc/nsswitch.conf passwd: compat group: compat shadow: compat hosts: files mdns4_minimal [NOTFOUND=return] dns mdns4 networks: files protocols: db files services: db files ethers: db files rpc: db files netgroup: nis But if fact it is not. user@test ~ping test PING localhost (127.0.0.1) 56(84) bytes of data. [skip] Pinging is ok. user@test ~host test test.mydomain.com has address xx.xxx.161.201 But pure I suspect that NetworkManager might cause this misbehavior, but don't know where to start to check it. Any thoughts, suggestions?

  Read the article

• Strange Apache Webdav situation (OSX Will connect, Ubuntu will not)

  - by mewrei

  So basically my situation is that I have an Apache 2.2 webserver running on Linux on another box, and I have it configured to serve up webdav. Now here's the weird part, I can access the server just fine on my Mac using the "Connect to Server" dialog (even moved like 5GB of files over the connection). On my Ubuntu desktop cadaver will connect as well and allow me to browse. However when I try to use Xmarks (BYOS Edition) or the GNOME "Connect to Server" dialog, it gives me a 403 Forbidden error. My server does digest authentication if that makes any difference. Here's part of my apache2.conf file <VirtualHost *:80> DocumentRoot "/path" <Directory "/path"> Dav on AuthType Digest AuthName iTools AuthDigestDomain "/" AuthUserFile /path/to/WebDavUsers Options None AllowOverride None <LimitExcept GET HEAD OPTIONS> require valid-user </LimitExcept> Order allow,deny Allow from All </Directory> <Directory "/path/*/Public"> Options +Indexes </Directory> <Directory "/path/user"> <LimitExcept GET HEAD OPTIONS> require user user </LimitExcept> </Directory> </VirtualHost>

  Read the article

• Deploying concrete5 on nginx

  - by Nithin

  I have a concrete5 site that works 'out of the box' in apache server. However I am having a lot of trouble running it in nginx. The following is the nginx configuration i am using: server { root /home/test/public; index index.php; access_log /home/test/logs/access.log; error_log /home/test/logs/error.log; location / { # First attempt to serve request as file, then # as directory, then fall back to index.html try_files $uri $uri/ index.php; # Uncomment to enable naxsi on this location # include /etc/nginx/naxsi.rules } # pass the PHP scripts to FastCGI server listening on unix socket # location ~ \.php($|/) { fastcgi_pass unix:/tmp/phpfpm.sock; fastcgi_split_path_info ^(.+\.php)(/.+)$; fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name; fastcgi_param PATH_INFO $fastcgi_path_info; include fastcgi_params; } location ~ /\.ht { deny all; } } I am able to get the homepage but am having problem with the inner pages. The inner pages display an "Access denied". Possibly the rewrite is not working, in effect I think its querying and trying to execute php files directly instead of going through the concrete dispatcher. I am totally lost here. Thank you for your help, in advance.

  Read the article

• Apache is spawning more and more processes!!

  - by erotsppa

  We have a LAMP setup that is working prety good for half a year. All of a sudden today the apache server (mysql servers are not on this box) started to die. It seems to have started to spawn more and more processes over time. Eventually it will consume all the memory and the server would just die. We are using prefork. In the mean time what we are doing is just added more ram and increased the MaxClients and ServerLimit parameter to 512. We're just prolonging the crash. The number still goes up slowly. Maybe in a day, it would reach that limit. What is going on? We only have around 15-20 request per second. We have 1Gb memory and it's not half used, there's no swapping going on. Why is apache creating more and more processes? It's almost like theres a leak somewhere! The database boxes are fine, they are not causing a delay to requests. We tested some queries everything is quick!

  Read the article

• Concerns about Apache per-Vhost logging setup

  - by etienne

  I'm both senior developer and sysadmin in my company, so i'm trying to deal with the needs of both activities. I've set up our apache box, wich deals with 30-50 domains atm (and hopefully will grow larger) and hosts both production and development sites, with this directory structure: domains/ domains/domain.ext/ #FTPS chroot for user domain.ext domains/domain.ext/public #the DocumentRoot of http://domain.ext domains/domain.ext/logs domains/domain.ext/subdomains/sub.domain.ext domains/domain.ext/subdomains/sub.domain.ext/public #DocumentRoot of http://sub.domain.ext Each domain.ext Vhost runs with his dedicated user and group via mpm-itk, umask being 027, and the logs are stored via a piped sudo command, like this: ErrorLog "| /usr/bin/sudo -u nobody -g domain.ext tee -a domains/domain.ext/logs/sub.domain.ext_error.log" CustomLog "| /usr/bin/sudo -u nobody -g domain.ext tee -a domains/domain.ext/logs/sub.domain.ext_access.log" combined Now, i've read a lot about not letting the logs out of a very restricted directory, but the developers often need to give a quick look to a particular subdomain error log, and i don't really want to give them admin rights to look into /var/logs. Having them available into the ftp account is REALLY handy during development stages. Do you think this setup is viable and safe enough? To me it is apparently looking good, but i'm concerned about 3 security issues: -is the sudo pipe enough to deal with symlink exploits? Any catches i'm missing? -log dos: logs are in the same partition of all domains. got hundreds of gigs, but still, if one get disk-space dos'd, everything will break. Any workaround? Will a short timed logrotate suffice? -file descriptors limits: AFAIK the default limit for Apache on Ubuntu Server is currently 8192, which should be plenty enough to handle 2 log files per subdomain. Is it? Am i missing something? I hope to read some thoughts on the matter!

  Read the article

• Ubuntu lucid, maverick high iowait

  - by netom

  I'm using Ubuntu, and I've got the same problem with Lucid and Maverick. From time to time, especially a few minutes after boot, the iowait goes between 50-100% and the box is unusable. Everything that tries to access the disk freezes. I have the following setup: Hard disk: Model Family: Western Digital Caviar Green family Device Model: WDC WD15EADS-00P8B0 Serial Number: WD-WMAVU0391287 Firmware Version: 01.00A01 User Capacity: 1.500.301.910.016 bytes I have a quad core Intel Core2 Q6600 processor, and 4G of memory. When the high iowait occurs, usually 4 processes are active: kdmflush (two procs) jbd2/dm-0-8 jbd2/db-1-8 and a few more starving user processes of course. I know this from top and iotop. Any suggestions about why this is happening? There are a lot of q/a-s about linux and high iowait, but none of them helped so far, I even tweaked the hard disk not to park the head in every 8 seconds (Load cycle count is 50334!!!!! :o ), but nothing. Problem persists. Thank you in advance.

  Read the article

• Debugging a Drobo that chokes Windows 7x64 When Plugged In

  - by Pridkett

  I've had a love hate relationship with my Drobo for a long time. After two years of using it on a Linux box, I moved it over to a Windows 7 machine where it seemed to work just fine for a long time, but it was under very light usage. Mainly backups that never actually happened. Recently I began using it for additional backup services (through CrashPlan, which is great). This means the Drobo gets a lot more usage. Also it means that something interesting happens, the Drobo can choke my system on startup. Here's what I mean: Start computer without Drobo plugged in, CrashPlan and Drobo Dashboard services disabled: 105s Start computer with Drobo plugged in Crashplan disabled, Drobo Dashboard enabled: 250s (and 1 cpu at 100% for a very long time, drobo churning) Start computer with Drobo plugged in, CrashPlan and Drobo Dashboard disabled: 250s (1 cpu at 100% for a very long time, drobo churning) Start computer with Drobo plugged in, Crashplan and Drobo Dashboard enabled: 300s (1 cpu at 100% for a very long time, drobo churning) If I yank the USB plug on the Drobo the CPU usage goes down to nothing very quickly. The slow startup in the fourth scenario is because CrashPlan is trying desperately to load stuff up on the H: drive before it gives up, so I've disabled it for the time being. So here's my question: What the heck is going on when I plug the drobo in? I've fired up Process Explorer and see that the System process is hogging the CPU, specifically it's an ntoskrnl.exe/KdPollBreakIn thread that's going ape. Is this something that's wrong with Drobo? Windows? Any idea on how to find out? If it matters, here's tech info: Athlon 64x2 4400, 2GB RAM, Win7 Ultimate, Drobo USB (2x1TB, 2x320GB)

  Read the article

• Restrict SSH user to connection from one machine

  - by Jonathan

  During set-up of a home server (running Kubuntu 10.04), I created an admin user for performing administrative tasks that may require an unmounted home. This user has a home directory on the root partition of the box. The machine has an internet-facing SSH server, and I have restricted the set of users that can connect via SSH, but I would like to restrict it further by making admin only accessible from my laptop (or perhaps only from the local 192.168.1.0/24 range). I currently have only an AllowGroups ssh-users with myself and admin as members of the ssh-users group. What I want is something that works like you may expect this setup to work (but it doesn't): $ groups jonathan ... ssh-users $ groups admin ... ssh-restricted-users $ cat /etc/ssh/sshd_config ... AllowGroups ssh-users [email protected].* ... Is there a way to do this? I have also tried this, but it did not work (admin could still log in remotely): AllowUsers [email protected].* * AllowGroups ssh-users with admin a member of ssh-users. I would also be fine with only allowing admin to log in with a key, and disallowing password logins, but I could find no general setting for sshd; there is a setting that requires root logins to use a key, but not for general users.

  Read the article

< Previous Page | 400 401 402 403 404 405 406 407 408 409 410 411  | Next Page >