Search Results

Search found 4893 results on 196 pages for 'expect'.

Page 127/196 | < Previous Page | 123 124 125 126 127 128 129 130 131 132 133 134  | Next Page >

• Buying an old laser printer -- what will need to be replaced?

  - by marienbad

  Hi all -- as you can see I'm new. I do IT and wiring for a small local shop but I never deal with printers. I do a LOT of printing, and I'd like to stop spending as much money on it. On my local CL, there is an HP 8100DN (duplex network) printer for a very good price (and the toner is a quarter-cent per page). It has printed 200,000 pages and I don't yet know anything else about it. The model was released in 1999. So my questions: What are the parts that tend to need service on laser printers? On ebay, I see fusers, rollers, DC power boards, and motors. What would you expect to replace soon at 200,000 pages? Are there any good "tests" to find out if certain parts are near failure? Do you have anything to say about the HP 8100 specifically? The bottom line for me is that if there's any chance of repairs costing more than $100, it's not worth it for me.

  Read the article

• mysqldump --where with = operator doesn't get all rows = - Help!

  - by JonathanLIVE

  I have a situation with a particular table that now thinks it contains 4 Petabytes of data. I know that sounds cool, but I assure you, it is only on a 60GB partition. This table has 9 fields in it. One of them is a domain_id field. It is the best field to identify the rows by, as there are only approximately 6300 of them. The only other field option to match has over 2million records, and thats just more difficult. I cannot do a straight mysqldump because it will attempt to output all 4PB of data and fill the drive long before it gets close to that, so I need to surgically remove the good stuff, destroy the db, and recreate it. I believe if I can do a dump for each domain_id record, then I will get most of the usable data out of it. This is what I am trying to use: mysqldump -u root --skip-opt -q --no-create-info --skip-add-drop-table --max_allowed_packet=1000000000 database table --where="domain_id=10" domains10.sql Using this I expect every row with the domain_id 10 to be exported. However, when I check the export, I am only getting 1 row, when however I look at the db, there are many many rows. It is as though the operator just finds one, then gives up. I have tried various operators. Using the < or I am able to get more of the data, but the export stops short at certain rows where the data has been compromised. With over 6000 to go through, I can't narrow down which rows are being affected in the export easily enough. So, what I need is an operator that will basically do what I thought = would do, simply give me an export of all records that match the specific field. Also note, the only way I got this DB even accessible is through an innodb force recovery 3. So I need to get this right, because after this is done, I have to drop the db in order to make mysql functional again. Looking forward to any helpful answers.

  Read the article

• Run Python script at startup using upstart

  - by MarcusMaximus

  I'm trying to create an upstart script to run a python script on startup. In theory it looks simple enough but I just can't seem to get it to work. I'm using a skeleton script I found here and altered. description "Used to start python script as a service" author "Me <[email protected]>" # Stanzas # # Stanzas control when and how a process is started and stopped # See a list of stanzas here: http://upstart.ubuntu.com/wiki/Stanzas#respawn # When to start the service start on runlevel [2345] # When to stop the service stop on runlevel [016] # Automatically restart process if crashed respawn # Essentially lets upstart know the process will detach itself to the background expect fork # Start the process script exec python /usr/local/scripts/script.py end script The test script I want it to run is currently a simple python script that runs without any issue when run from a terminal. #!/usr/bin/python2 import os, sys, time if __name__ == "__main__": for i in range (10000): message = "UpstartTest " , i , time.asctime() , " - Username: " , os.getenv("USERNAME") #print message time.sleep(60) out = open("/var/log/scripts/scriptlogfile", "a") print >> out, message out.close() The location/var/log/scripts has permissions 777 The file /usr/local/scripts/script.py has permissions 775 The upstart script /etc/init.d/pythonupstart.conf has permissions 755

  Read the article

• Bizarre client IP switch-up on VPN

  - by B. VB.

  Let A.B.C.D be the public IP of my VPN server. Let W.X.Y.Z be the IP of the client before it connects to the VPN. My VPN server's IP address on the LAN in 10.8.0.1, and the client is 10.8.0.6. I also run a webserver on the same machine hosting the VPN. On it is a simple webpage that performs the exact same thing as whatismyip.org (i.e., simply prints the IP of the requester) Let me illustrate the scenario for you. In a Chrome window I have three tabs, what I have in parenthesis is the URL: Tab 1 (http://whatismyip.org): A.B.C.D This is what I expect to see. It's the public IP of the VPN server. Tab 2 (http://10.8.0.1): 10.8.0.6 ok, looks expected. They are behind the same LAN now. Tab 3 (http://A.B.C.D) W.X.Y.Z WTF?? Basically, if I access the webserver while tunneled, in shows the IP address of my machine PRIOR to tunelling! Remember, tab2 and tab3 are the same webpage. Why does Tab3 not show the client IP as it's own IP (i.e., show A.B.C.D)??? I hope this question is clear, thanks in advance!

  Read the article

• Apache Virtualhost entry with Windows hostname

  - by gshauger

  I have a Windows Domain Controller and we use it for DNS for our internal network. I have an Ubuntu box with an IP address of 172.16.34.149. Within the Windows DNS I created the forward and reverse lookup entries for the name Endymion. Naturally when ever I FTP/SSH/HTTP/etc to the hostname Endymion it resolves correctly to my Ubuntu box. I wanted to do some web development on this box for an existing site. There were problems when I placed the website in a subfolder of /var/www/. Let's just say it was in folder /var/www/projectx/. The issue involved the incorrect resolution of non-relative urls. So I figure I could create a new DNS entry for the hostname projectx. Sure enough when I FTP/SSH/HTTP/etc to the hostname projectx it takes me to the same ubuntu box as the hostname Endymion...this is what I would expect. I now have two hostnames for the same box. I then create a Virtualhost entry in httpd.conf that looks like the following: <VirtualHost *:80> DocumentRoot /var/www/projectx ServerName projectx ServerAlias projectx </VirtualHost> Sure enough when I go to a browser and type in http://projectx/ it takes me to the correct subfolder. Everything works!!! Not so fast. I then go to http://endymion/ and instead of taking me to /var/www/ it takes me to /var/www/projectx/ Clearly I'm missing something. Help please! ;)

  Read the article

• Synchronizing the SamAccountName Property using Windows Azure Active Directory Sync Tool

  - by pk.

  Using this official documentation as a guide, I would expect the SamAccountName property to sync from my on-premise AD to Office 365. I think that it used to do exactly that, but now it seems that it doesn't so much sync the attribute as it does create an entirely new, unlinked value and store it in Office 365. This has caused some minor issues for me (broken scripts, annoying permissions management, etc.) and may be part of a more major issue regarding ADFS authentication. On-Premise PS C:\Windows\system32> Get-ADUser jdoe -Properties SamAccountName | fl SamAccountName SamAccountName : jdoe Office 365 Sync'ed Objects PS C:\Windows\system32> Get-Mailbox jdoe | fl SamAccountName SamAccountName : $1A7H20-K1LCOJFFBHGS I understand how to work around this issue in my scripts -- there exists the ImmutableId property which can be mapped back to the on-premise GUID. As far as the issue I'm having with ADFS, I'm less certain how to proceed and if this is causing my issues. At this point I really would just like some verification that I'm not crazy and that this used to be sync'ed at some point in the past and that Office 365 broke it relatively recently. I also think that MS documentation should perhaps be updated to exclude SamAccountName from the list of synchronized properties on the page I linked.

  Read the article

• MySQL : table organisation for very large sets with high update frequency

  - by Remiz

  I'm facing a dilemma in the choice of my MySQL schema application. So before I start here is a picture extremely simplified of my database : Schema here : http://i43.tinypic.com/2wp5lxz.png In one sentence : for each customer, the application harvest text data and attached tags to each data collected. As approximation of the usage of each table, here is what I expect : customer : ~5000, shouldn't grow fast data : 5 millions per customer, could double or triple for big customers. tag : ~1000, quite fixed size data_tag : hundred of millions per customer easily. Each data can be tagged a lot. The harvesting process is permanent, that means that around every 15 minutes new data come and are tagged, that require a very constant index refreshing. A lot of my queries are a SELECT COUNT of DATA between specific DATES and tagged with a specific TAG on a specific CUSTOMER (very rarely it will involve several customers). Here is the situation, you can imagine with this kind of volume of data I'm facing a challenge in term of data organization and indexing. Again, it's a very minimalistic and simplified version of my structure. My question is, is it better: to stick with this model and to manage crazy index optimization ? (which involves potentially having billions of rows in the data_tag table) change the schema and use one data table and one data_tag table per customer ? (which involves having 5000 tables on my database) I'm running all of this on a MySQL 5.0 dedicated server (quad-core, 8Go of ram) replicated. I only use InnoDB, I also have another server that run Sphinx. So knowing all of this, I can't wait to hear your opinion about this. Thanks.

  Read the article

• Nexenta, NFS and LOCK_EX

  - by Givre

  I'm currently using an LAMP architecture and I expect a big problem :( I have several http web server using PHP5. All are mounting via NFS (v3) the directory for all the hosted websites. The file server is running the Nexenta Storage Appliance using ZFS . The problem is all the NFS client trying to write something in a file over the NFS get this problem : This is inside the apache2 process: open("/nfs/website1/file.txt", ORDWR|OCREAT, 0600) = 11647 fstat(11647, {stmode=SIFREG|0600, st_size=23754, ...}) = 0 flock(11647, LOCK_EX And the process never get the LOCK and keep waiting for... always. The effect? All the apache2 procces get used and waiting.. my severs can't still proccess the others requests because there is no more proccess available. I don't now where to find a solution.. for me it.'s on the NFS server side.. but wich configuration is wrong or missing ? How can I find what is wrong? If you need more information about the configuration, just ask me what can help you more :)

  Read the article

• Access 2010 datasheet view only/relationships unavailable

  - by Luis

  I'm relatively new to MS Access in general and just started working with Access 2010. I've created a new web database with a few tables that I need to relate. First problem: For the life of me, I can't view anything in any view other than datasheet view; everywhere I would expect to be able to change the view, only datasheet view is available. Second problem: I can't change the primary key(s). Presumably I would be able to do this if I could get out of datasheet view and into design view. Third problem: The 'Relationships' button is greyed out. I know these appear to be really simple things but I've been looking for much more time than I'd like to admit trying to figure out how to get unstuck. Update: It would appear that this is happening because it is a 'web database' as I've been able to do all of the above in a new regular database. With this in mind let me ask a different question: Am I able to add relationships and change primary keys in a web database? If so how? More generally, what is the point of a web database?

  Read the article

• Can I install windows on an SSD and access data from my old windows HDD?

  - by nzifnab

  I purchased new computer components, switching my hardware from AMD and Radeon to Intel and Nvidia. I kept components from my old computer like the powersupply and two HDDs. Everything appeared to install correctly and the system booted into the BIOS just fine (after a brief snafu with the CPU fan). My goal was to use the two harddrives and just be able to turn on the computer and load up my old windows install with all the files, programs, and documents. I expected to have to call Microsoft to re-register the windows install for the new hardware (since I had to do that last time I upgraded w/ the same windows version). When the computer attempts to boot into windows it briefly flashes a bluescreen and then restarts. System recovery gives a message something like "BadDriver Failover" something something. I assume this is because it's trying to use amd drivers for an intel chipset (or something...?) and I've been as-yet unsuccessful in getting it to boot into my old windows partition. SO! I decided eff it, maybe I'll go visit my nearest Micro Center and buy a 200 GB SSD, install windows onto that, and then... be able to access just the contents of both of my other harddrives? I don't intend on running any of the programs but there were some saved files I would like to salvage from the 500 GB harddrive, The 1.5 TB harddrive only had files on it, no OS or applications so I'd also expect to still be able to access it. Is this possible? Can I install the SSD and only format/install windows onto it and still access the contents from my two transferred-over drives?

  Read the article

• CentOS server. What does it mean when the total used RAM does not equal the sum of RES?

  - by Michael Green

  I'm having a problem with a virtual hosted server running CentOS. In the past month a process (java based) that had been running fine started having problems getting memory when the JVM was started. One strange thing I've noticed is that when I start the process, the PID says it is using 470mb of RAM while the 'used' memory immediately drops by over a 1GB. If I run 'top', the total RES used across all processes falls short of the 'used' listed at the top by almost 700mb. The support person says this means I have a memory leak with my process. I don't know what to believe because I would expect a memory leak to simply waste the memory the process is allocated not to consume additional memory that doesn't show up using 'top'. I'm a developer and not a server guy so I'm appealing to the experts. To me, if the total RES memory doesn't add up to the total 'used' it indicates that something is wrong with my virtual server set-up. Would you also suspect a memory leaking java process in this case? If I use free before: total used free shared buffers cached Mem: 2097152 149264 1947888 0 0 0 -/+ buffers/cache: 149264 1947888 Swap: 0 0 0 free after: total used free shared buffers cached Mem: 2097152 1094116 1003036 0 0 0 -/+ buffers/cache: 1094116 1003036 Swap: 0 0 0 So it looks as though the process is using (or causing to be used) nearly 1GB of RAM. Since the process (based on top is only using 452mb, does that mean that the kernal is all of a sudden using an additional 500mb?

  Read the article

• How to deal with ssh's "WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED!"?

  - by Vi.

  I often need to login to multiple remote stations that are just placed to the same static IPs for me. SSH complains about changed keys in this case: $ ssh [email protected] @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ @ WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED! @ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ ... Offending RSA key in /home/vi/.ssh/known_hosts:70 ... I usually just run vim /home/vi/.ssh/known_hosts +70, dd wq and re-run the SSH command. How to do it simpler? Requirements: The warning should be displayed, and not like this: The authenticity of host '172.1.2.3 (172.1.2.3)' can't be established. It is easy to accept the key change. I expect something like this: $ ssh [email protected] @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ @ WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED! @ @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ ... The fingerprint for the RSA key sent by the remote host is 82:cd:be:7a:ae:1b:91:2c:23:c1:74:4d:8a:38:10:32. Change the host key in /home/vi/.ssh/known_hosts (yes/no)? yes Warning: Changed host key for '172.1.2.3' (RSA) in the list of known hosts. [email protected]'s password: Simple and differs from usual "The authenticity of host can't be established." message.

  Read the article

• What's the safest way to kick off a root-level process via cgi on an Apache server?

  - by MartyMacGyver

  The problem: I have a script that runs periodically via a cron job as root, but I want to give people a way to kick it off asynchronously too, via a webpage. (The script will be written to ensure it doesn't run overlapping instances or such.) I don't need the users to log in or have an account, they simply click a button and if the script is ready to be run it'll run. The users may select arguments for the script (heavily filtered as inputs) but for simplicity we'll say they just have the button to choose to press. As a simple test, I've created a Python script in cgi-bin. chown-ing it to root:root and then applying "chmod ug+" to it didn't have the desired results: it still thinks it has the effective group of the web server account... from what I can tell this isn't allowed. I read that wrapping it with a compiled cgi program would do the job, so I created a C wrapper that calls my script (its permissions restored to normal) and gave the executable the root permissions and setuid bit. That worked... the script ran as if root ran it. My main question is, is this normal (the need for the binary wrapper to get the job done) and is this the secure way to do this? It's not world-facing but still, I'd like to learn best practices. More broadly, I often wonder why a compiled binary is more "trusted" than a script in practice? I'd think you'd trust a file that was human-readable over a cryptic binaryy. If an attacker can edit a file then you're already in trouble, more so if it's one you can't easily examine. In short, I'd expect it to be the other way 'round on that basis. Your thoughts?

  Read the article

• How should an experienced Windows SysAdmin learn Linux? [closed]

  - by Systemspoet

  I have a new hire starting in a few weeks who is an experienced Windows SysAdmin. I think he's fairly senior on the Windows side, with a pretty deep AD understanding and experience with Exchange 2007, 2010, and exchange migrations. He's done a little PowerShell but I suspect more of the "run this command to do this" variety then "write a script to do this" sort. However, we are a mixed shop and (he knows this) I expect him to become a reasonably competent Linux SysAdmin over time. I'm looking for good starting points to bring him along. I have over ten years of Linux/UNIX experience, so it all sort of seems intuitive to me, but I've been thinking about the toolkit you actually need to be productive in the Linux CLI world. Just to be able to use the machines at all, off the top of my head... vi Basic CLI stuff -- move around, rename files, copy files, tar, gzip, changing passwords, finding relevant manpages, keep track of where you are, find things in your history, etc, etc. More advanced things that I take for granted but are actually pretty hard -- doing things with 'find', extracting relevant text via 'awk' and/or 'cut', knowing when to use 'grep' and when to use 'grep -e' or 'egrep'. Distribution specific stuff... compiling software, rpm, yum, apt-get, you name it. This all seems pretty basic to me, but when I think back to 1995 when I was first learning my way, some of those things took me years to master. So my question is -- where should I send him to pick up those skills? I'm not just thinking of classes, but rather also websites and books? Where do you all suggest as a starting point for picking up Linux skills?

  Read the article

• Why does my computer not always automatically request a new DHCP lease when switching networks?

  - by KingJ

  I've noticed an interesting problem recently where my laptop will not automatically refresh my DHCP lease when changing between wireless networks. Instead, it will attempt to use the existing lease for the new network. Of course, since each network has different settings any connections will fail when using the old lease. While Windows will eventually request a new DHCP lease, it's often much faster for me to release the old lease and request a new, correct, lease. It puzzles me as to why this is not automatically done when associating to the network? One possible cause is that the laptop is put to sleep while connected to one network, then woken when in range of the other network. Nevertheless, I would expect that the lease would be renewed when the laptop associates to the other network rather than attempting to use the old lease. It should be noted that the two networks are completely different, both in terms of size, number of APs, network settings, SSIDs etc. One network is my home network and the other one is the campus-wide network at my University.

  Read the article

• 2.6.9 Kernel on virtual server (non upgradable) - any expected problems?

  - by chris_l

  Hi, I'm considering to rent a virtual server (for me personally). The product I'm currently looking at offers IMO fair pricing, very good hardware etc. The only problem is, that I won't be able to do an upgrade to a newer kernel than 2.6.9 (running Debian Etch). Also, I can't install my own kernel modules. (The server runs with Virtuozzo, so as far as I understand it, it just does some chroot instead of a real virtualization (?)) I want to run GlassFish, Postgres, Subversion, Trac and maybe some other things on it. It will also have to employ a firewall, and provide OpenSSL for https. Ideally, it would also be able to do AIO (asynchronous IO), which could speed up some server I/O. Should I expect problems with that old kernel version, in conjunction with the software I want to install (I'd like to use current versions of the software)? One thing I already found out, is that you can't do everything with iptables, since some kernel modules are missing/things are not build into the kernel. GlassFish v3 appears to run fine at first glance. I was able to test the server for a few hours. Installing my whole setup wasn't feasible in that time, but what I can say is, that it's amazingly fast for an entry-level vserver, especially hard disk and network performance (averaging at ca. 400MBit/s). So if the kernel won't be a problem, I'd really like to take it. Thanks, Chris PS Exact kernel version: 2.6.9-023stab051.3-smp

  Read the article

• Linux Firefox copy/paste issue

  - by Daniel

  I'm using Firefox 15.0.1 on Fedora 17 without running gnome or kde. The problem I'm having is that whenever I select text outside of Firefox, for instance in xterm, the middle mouse button doesn't copy it inside Firefox, for instance in a text area, but rather brings up a context menu. A related problem is that whenever I middle click inside Firefox, for instance in a text input, the middle mouse button brings up a menu when I'd like it to paste. Even if I select Paste in the menu not the selected text (from outside Firefox) gets pasted but the last selected text inside Firefox. In about:config I tried "middlemouse.paste true" and also "middlemouse.paste false" together with the add-on Auto Copy but no combination worked. A middle mouse click always brings up a context menu. But the Auto Copy did help with automatically copying selected text to the clipboard. With Auto Copy the only problem I still have is pasting by middle button. Follow up: somehow the problem solved itself. After removing Auto Copy firefox works as I expect it to (as any X application). I can't figure out why it was not doing it before, probably I was messing too much with about:config and not restarting frequently enough.

  Read the article

• Computer is dying--what should I be looking for?

  - by Will

  Okay, I'm a bit knowledgeable with pooters and such, but i'm confused. My computer is dying slowly, and I'm not sure what part is causing this. Computer details: Vista, dell machine, intel Q6600, 2.4 Core Duo (quad core), standard memory and drive (unknown manufacturer). Symptoms: I would best describe the symptoms as memory corruption. After a couple days on, I start getting applications crashing or failing to open for a lack of "resources". Sounds are corrupted. Onscreen text gets corrupted; the characters of text are garbled, not the pixels on the screen. Video memory seems untouched as I haven't seen any misplaced pixels. Recently I've lost files on disk. I've also experienced errors reporting a supposed lack of disk space, even though I have fifty gigs free. There was one point where I couldn't get to the POST when booting up. After I cleaned everything (see next) this hasn't happened. Diagnostic steps: First thing I did was clean the case. There was a lot of dust buildup on heatsinks, so I cleaned all that up. No help. Next, I disconnected and reconnected everything, from power cables to memory (did not reseat cpu). No change. Last, I ran the standard vista memory diagnostics and ran checkdisk. Both reported no errors found. I have not run any POST tests, now that I think about it. I'm at a loss at this point. Disk appears fine, memory too. I'd expect motherboard issues to result in the thing not booting up, yet it does every time. What should I be looking at? What more can I do?

  Read the article

• What are possible results/side effects if replication between DC's in a Windows domain is unable to occur?

  - by hydroparadise

  There's plenty of administration literature out there how to properly manage Windows servers. But in dealing with real life, things don't always occur like you want them to. In Microsoft's Windows Server 2003 Administrator's Companion, out of 1400+ pages, theres only one page that I could find when it comes up setting up additional domain controlers. They make it sound seemless and don't reveal a whole lot on what happens if "peer" DC's are unable to replicate. Down to the specific issue at hand, we had a DC go down about a month ago due to a bad RAID controller. There was nothing critical that waranted imediate attention, so bringing it back up got put on the back burner. A month later, we get the DC back up and running and everyting seemed ok. The next day, nobody is able to logon complaining that the "user does not exist" or "unable to establish a trust relationship". Knowing that I had just put the downed DC back on the network, I immediately took it back off the network and had everybody restart the workstations. After that, exchange was fine, shares became available, and everybody was able to log in. After doing some event log swimming, it would appear that everything started due to replication issues on the SYSVOL. I've read where you can force replication, but that would mean putting it back on the network. I am afraid to put the DC back on the network in fear that something else could go wrong. So, what other issues could one expect to run into where two DC's are unreplicated for over a month?

  Read the article

• How to get desired FireFox last tab behavior?

  - by JustJeff

  All tabs should be the same; so if any of the have a 'close' button, they all should, including the last tab. I see no reason that a tab's close button should suddenly vanish simply b/c that tab has become the last one open. If I have N tabs open, and park the mouse over the left-most tab's close button, this vanishing close button trick means now I have to make a large mouse move to get to the app's close button. Unsat. Mouse moves = too many milliseconds wasted. Closing the last tab should NOT take me to my home page, or any other page whatsoever. I want the browser to close with the last tab. I do not expect or want "new tab" behavior when I click a Close button. Now, I've gone into about:config and played with browser.tabs.closeWindoWithLastTab, but this setting oversteps its purpose; while it does make the browser close, for some inexplicable reason, it also suppresses the last tab's close button! I have tried the "last tab close button" add-on, and while this does restore the close button, the add-on oversteps by taking the liberty of turning closeWindowWithLastTab off. Is there some way out of this pickle? Is it too hard to just code things to provide simple, orthogonal actions, so that everybody can config the UI to their liking, and not just to a few pre-fab configurations that the developers think everyone should like? Btw, FF 13.0.1 on ms windows

  Read the article

• Ubuntu's garbage collection cron job for PHP sessions takes 25 minutes to run, why?

  - by Lamah

  Ubuntu has a cron job set up which looks for and deletes old PHP sessions: # Look for and purge old sessions every 30 minutes 09,39 * * * * root [ -x /usr/lib/php5/maxlifetime ] \ && [ -d /var/lib/php5 ] && find /var/lib/php5/ -depth -mindepth 1 \ -maxdepth 1 -type f -cmin +$(/usr/lib/php5/maxlifetime) ! -execdir \ fuser -s {} 2> /dev/null \; -delete My problem is that this process is taking a very long time to run, with lots of disk IO. Here's my CPU usage graph: The cleanup running is represented by the teal spikes. At the beginning of the period, PHP's cleanup jobs were scheduled at the default 09 and 39 minutes times. At 15:00 I removed the 39 minute time from cron, so a cleanup job twice the size runs half as often (you can see the peaks get twice as wide and half as frequent). Here are the corresponding graphs for IO time: And disk operations: At the peak where there were about 14,000 sessions active, the cleanup can be seen to run for a full 25 minutes, apparently using 100% of one core of the CPU and what seems to be 100% of the disk IO for the entire period. Why is it so resource intensive? An ls of the session directory /var/lib/php5 takes just a fraction of a second. So why does it take a full 25 minutes to trim old sessions? Is there anything I can do to speed this up? The filesystem for this device is currently ext4, running on Ubuntu Precise 12.04 64-bit. EDIT: I suspect that the load is due to the unusual process "fuser" (since I expect a simple rm to be a damn sight faster than the performance I'm seeing). I'm going to remove the use of fuser and see what happens.

  Read the article

• SSH & SFTP: Should I assign one port to each user to facilitate bandwidth monitoring?

  - by BertS

  There is no easy way to track real-time per-user bandwidth usage for SSH and SFTP. I think assigning one port to each user may help. Idea of implementation Use case Bob, with UID 1001, shall connect on port 31001. Alice, with UID 1002, shall connect on port 31002. John, with UID 1003, shall connect on port 31003. (I do not want to lauch several sshd instances as proposed in question 247291.) 1. Setup for SFTP: In /etc/ssh/sshd_config: Port 31001 Port 31002 Port 31003 Subsystem sftp /usr/bin/sftp-wrapper.sh The file sftp-wrapper.sh starts the sftp server only if the port is the correct one: #!/bin/sh mandatory_port=3`id -u` current_port=`echo $SSH_CONNECTION | awk '{print $4}'` if [ $mandatory_port -eq $current_port ] then exec /usr/lib/openssh/sftp-server fi 2. Additional setup for SSH: A few lines in /etc/profile prevents the user from connecting on the wrong port: if [ -n "$SSH_CONNECTION" ] then mandatory_port=3`id -u` current_port=`echo $SSH_CONNECTION | awk '{print $4}'` if [ $mandatory_port -ne $current_port ] then echo "Please connect on port $mandatory_port." exit 1 fi fi Benefits Now it should be easy to monitor per-user bandwidth usage. A Rrdtool-based application could produce charts like this: I know this won't be a perfect calculation of the bandwidth usage: for example, if somebody launches a bruteforce attack on port 31001, there will be a lot of traffic on this port although not from Bob. But this is not a problem to me: I do not need an exact computation of per-user bandwidth usage, but an indicator that is approximately correct in standard situations. Questions Is the idea of assigning one port for each user is a good one? Is the proposed setup an reliable one? If I have to open dozens of ports for many users, should I expect a performance drawback? Do you know a rrdtool-based application which could make the chart above?

  Read the article

• Video adapter problem on new motherboard

  - by Dmitriy Matveev

  Something bad happened to my PC several days ago. I wasn't near my PC, so I have no idea about what happened it was just halted. When I tried to boot it was entering infinite loops of a few seconds power up (cpu, hdds, etc) and then power down. I've tried to boot the system with no additional components (that includes hdd, video, network adapters and even memory) connected and that didn't solved the problem, so I've made a decision that most likely it's either some a problem with power unit or with motherboard. I've tried to replace my power unit by another one (which was expected to be working) and the problem didn't resolved again. I've bought a new MB (ASUS P5KPL-AM SE) and tried to get it running with my old CPU and memory (I hope it's still alive). Since this MB include on-board video I've tried to run the PC without installation of mine video. The PC wasn't running and the BIOS was beeping one long signal following by two short (Does it means a video problem?). After that I've installed my video adapter to PCI-E slot and tried to boot the system again and the BIOS was beeping the same. I don't get it. I may expect some problem with CPU and/or memory since I don't know what happened to my PC (maybe some power failure or something different), but not with video and not with on-board video on newly bought MB. How can I understand what's wrong with my system now?

  Read the article

• Why doesn't the value in /proc/meminfo seem to map exactly to the system RAM?

  - by Eric Asberry

  The values in /proc/meminfo for MemTotal don't make sense. As a human, eyeballing it, it seems to roughly correspond to the installed RAM, but for using it to display the installed RAM from an automated utility it appears to be inexact, and inconsistent. For a system with 1G of RAM, I would expect the MemTotal line to have a value of 1048576 - 1024*1024. But instead, I'm seeing 1029392. On another 4G box, I'm seeing 3870172, which is not a multiple of 1024, and it's not even close to 1029392*4. On an 8G box, I get 8128204, which again seems to have no correlation to the other values, nor is it a multiple of 1024. I'm trying to use this information to report the RAM on a status web page. My work-around is to just "round" it to the nearest 1G multiple, but I'd like to understand why these values seem inconsistent and don't match my expectations. Can somebody fill me in on what I'm missing here? EDIT: To expand on the accepted answer below.... The reference can be found here. Also of interest to me from that page, which explains the inconsistency, is this bit: meminfo: Provides information about distribution and utilization of memory. This varies by architecture and compile options. ...

  Read the article

• Which free RDBMS is best for small in-house development?

  - by Nic Waller

  I am the sole sysadmin for a small firm of about 50 people, and I have been asked to develop an in-house application for tracking job completion and providing reports based on that data. I'm planning on building it as a web application. I have roughly equal experience developing for MySQL, PostgreSQL, and MSSQL. We are primarily a Windows-based shop, but I'm fairly comfortable with both Windows and Linux system administration. These are my two biggest concerns: Ease of managability. I don't expect to be maintaining this database forever. For the sake of the person that eventually has to take over for me, which database has the lowest barrier to entry? Data integrity. This means transaction-safe, robust storage, and easy backup/recovery. Even better if the database can be easily replicated. There is not a lot of budget for this project, so I am restricted to working with one of the free database systems mentioned above. What would you choose?

  Read the article

< Previous Page | 123 124 125 126 127 128 129 130 131 132 133 134  | Next Page >