Search Results

Search found 2282 results on 92 pages for 'filesystem'.

Page 77/92 | < Previous Page | 73 74 75 76 77 78 79 80 81 82 83 84  | Next Page >

• setup lowcost image storage server with 24x SSD array to get high IOPS?

  - by Nenad

  I want to build let's name it a lowcost Ra*san which would host for our social site the images (many millions) we have 5 sizes of every photo with 3 KB, 7 KB, 15 KB, 25 KB and 80 KB per Image. My idea is to build a Server with 24x consumer 240 GB SSD's in Raid 6 which will give me some 5 TB Disk space for the photo storage. To have HA I can add a 2nd one and use drdb. I'm looking to get above 150'000 IOPS (4K Random reads). As we mostly have read access only and rarely delete photos i think to go with consumer MLC SSD. I read many endurance reviews and don't see there a problem as long we don't rewrite the cells. What you think about my idea? - I'm not sure between Raid 6 or Raid 10 (more IOPS, cost SSD). - Is ext4 OK for the filesystem - Would you use 1 or 2 Raid controller, with Extender Backplane If anyone has realized something similar i would be happy to get Real World numbers. UPDATE I have buy 12 (plus some spare) OCZ Talos 480GB SAS SSD Drive's they will be placed in a 12-bay DAS and attached to a PERC H800 (1GB NV Cache, manufactured by LSI with fastpath) Controller, I plan to setup Raid 50 with ext4. If someone is wondering about some benchmarks let me know what you would like to see.

  Read the article

• Heartbeat/DRBD failover didn't work as expected. How do I make the failover more robust?

  - by Quinn Murphy

  I had a scenario where a DRBD-heartbeat set up had a failed node but did not failover. What happened was the primary node had locked up, but didn't go down directly (it was inaccessible via ssh or with the nfs mount, but it could be pinged). The desired behavior would have been to detect this and failover to the secondary node, but it appears that since the primary didn't go full down (there is a dedicated network connection from server to server), heartbeat's detection mechanism didn't pick up on that and therefore didn't failover. Has anyone seen this? Is there something that I need to configure to have more robust cluster failover? DRBD seems to otherwise work fine (had to resync when I rebooted the old primary), but without good failover, it's use is limited. heartbeat 3.0.4 drbd84 RHEL 6.1 We are not using Pacemaker nfs03 is the primary server in this setup, and nfs01 is the secondary. ha.cf # Hearbeat Logging logfacility daemon udpport 694 ucast eth0 192.168.10.47 ucast eth0 192.168.10.42 # Cluster members node nfs01.openair.com node nfs03.openair.com # Hearbeat communication timing. # Sets the triggers and pulse time for swapping over. keepalive 1 warntime 10 deadtime 30 initdead 120 #fail back automatically auto_failback on and here is the haresources file: nfs03.openair.com IPaddr::192.168.10.50/255.255.255.0/eth0 drbddisk::data Filesystem::/dev/drbd0::/data::ext4 nfs nfslock

  Read the article

• Copy past speed very slow for a large number of tiny files on Windows but not on linux

  - by Arno2501

  I've got this folder which contains 15'000 of tiny images (around 400 bytes each). If I copy past this folder on my laptop (Windows 7, i7 latest gen, superfast ssd) it takes about 30 seconds (yes for 7 megs !!!) the average transfer rate is 400 KBytes / second which is so slow. I mean my usual transfer rate is more like hundreds of MBytes per second !!! I get the same problem on my servers (Windows 2003, 2008 /r2) and on every Windows box that I could get my hands on. On the other hand if I do the same on a linux box (debian base, Ext3 FS) (which runs on the same SAN than all the windows servers I've tested) It's nearly instantaneous !!! I'm pretty sure the size / number of the files may stress such filesystem more than another but such differences !? Why is that ? Why is it so slow on the windows boxes (more that 30 sec for 7 MB) and so fast on the linux ones (one sec or so) (I mean this was not a hardlink that I've created it was a true copy). Is it a normal behaviour or something unusual ?

  Read the article

• Will these instructions work when turning of journaling on an ext4 SSD?

  - by snowlord

  I have an Acer Aspire One with an SSD for storage. I recently installed Ubuntu on it and chose ext4 for my filesystem. Then I read that journaling on an SSD isn't the best idea, so I will try to disable journaling and I have found these intstructions (from http://fenidik.blogspot.com/2010/03/ext4-disable-journal.html): # Create ext4 fs on /dev/sda10 disk mkfs.ext4 /dev/sda10 # Enable writeback mode. This mode will typically provide the best ext4 performance. tune2fs -o journal_data_writeback /dev/sda10 # Delete has_journal option tune2fs -O ^has_journal /dev/sda10 # Required fsck e2fsck -f /dev/sda10 # Check fs options dumpe2fs /dev/sda10 |more For more performance add fstab opions: data=writeback,noatime,nodiratime i.e: /dev/sda10 /opt ext4 defaults,data=writeback,noatime,nodiratime 0 0 I will use them on my boot partition. Are there any particularly bad parts here, or are there any missing steps? Will my boot partition be fit for being on an SSD after this? Or should I consider switching to ext2, or even reinstall it all and choose ext2 at partitioning time (I'd rather not though, since I've configured quite some stuff already)?

  Read the article

• Old scheduled task still being started, but can't find it.

  - by JvO

  System: Windows XP Home Summary: Some scheduled task is still being started by Windows, but I can't find it, nor determine where its configuration has been stored. This is turning into a mystery for me... I set up a Windows XP Home machine to run a task at 7:00 AM, using the Task manager. This was a clean install, no users defined, so you got straight to the desktop after starting the machine. The filesystem uses NTFS. Later on, I needed to introduce users, so I created one (named Sam) with administrator privileges. After this I noticed that the scheduled task failed, most likely due to privilege errors (i.e. can't write to a network drive). So I want to delete the old task, and add it again with the correct user credentials. However.... I can't find the old task!! I know it is still being executed at 7:00 AM, but there's no mention anywhere on the system of this task. I've looked in c:\windows\tasks for .job files, but there's only the "MP Scheduled Scan.job" from Security Essentials. I've searched the whole disk for mention of the batch file that is being run, but can't find it. So why is this old task still running, and more importantly, why can't I find it? Would it have something to do with introducing users on XP?

  Read the article

• Failing to load rootfs: Ubuntu 10 + grub2 + rootfs ext4 w/ RAID1

  - by James

  I am having problems booting a new Ubuntu 10 (server) install. My primary HD (/dev/sda) is laid out as follows: Device Boot Start End Blocks Id System /dev/sda1 * 1 18 144553+ 83 Linux <-- /BOOT /dev/sda2 19 182401 1464991447+ 5 Extended /dev/sda5 19 2207 17583111 fd Linux raid autodetect /dev/sda6 2208 11934 78132096 fd Linux raid autodetect <-- / (ROOTFS) /dev/sda7 11935 182401 1369276146 fd Linux raid autodetect The rootfs is part of a RAID1 (software) array (currently degraded): # cat /proc/mdstat Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10] md2 : active raid1 sda6[1] 78132032 blocks [2/1] [_U] The UUIDs for the partitions are as follows: # blkid /dev/sda1 /dev/sda1: UUID="b25dd301-41b9-4f4d-9b0a-0e31713dd74c" TYPE="ext2" # blkid /dev/sda6 /dev/sda6: UUID="af7b9ede-fa53-c0c1-74be-31ec752c5cd5" TYPE="linux_raid_member" # blkid /dev/md2 /dev/md2: UUID="a0602d42-6855-482f-870c-6f6ecdcdae3f" TYPE="ext4" Finally, I have my grub2 menuentry setup as follows: ### BEGIN /etc/grub.d/10_linux ### menuentry 'Ubuntu, with Linux 2.6.32-25-server' --class ubuntu --class gnu-linux --class gnu --class os { insmod ext2 insmod raid insmod mdraid set root='(hd0,1)' search --no-floppy --fs-uuid --set b25dd301-41b9-4f4d-9b0a-0e31713dd74c linux /vmlinuz-2.6.32-25-server root=UUID=a0602d42-6855-482f-870c-6f6ecdcdae3f ro nosplash noplymouth initrd /initrd.img-2.6.32-25-server } When I attempt to boot, grub loads OK, however I eventually get the following error message: Gave up waiting for root device. ALERT /dev/disk/by-uuid/a0602d42-6855-482f-870c-6f6ecdcdae3f does not exist. Dropping to a shell! If from the grub bootloader I open a grub command line, I can ls (hd0,) and it lists the correct partitions with the UUIDs as shown above - sda6 shows 'a0602d42-6855-482f-870c-6f6ecdcdae3f' (the RAID UUID). If I ls (md2)/ it properly lists all the files on the RAID1 filesystem (ext4) so it doesn't appear to be an issue accessing the raid device. Does anyone have any suggestions as to what the problem might be? I can't figure this one out.

  Read the article

• Apache DAV at `/` with normal hosting at `/foo` - how?

  - by mandrake

  Should I not be able to have a configuration where I serve SVN repos with SVNParentPath at <Location /> and then override DAV and host normal files using another location <Location /foo>? I wish to host my XSLT files on the same subdomain and still host repos at root. Of course, if I was to have a repo called foo, that would not be accessible, and that's ok. <VirtualHost *:80> ... #Host XSLT files here <Location /foo> DAV Off </Location> #Host my repos relative to root, such as /my_repo/ <Location /> DAV svn SVNParentPath "myrepos" SVNListParentPath on SVNIndexXSLT "/foo/my.xsl" ... </Location> </VirtualHost> But DAV SVN still looks for a repo: <?xml version="1.0" encoding="utf-8"?> <D:error xmlns:D="DAV:" xmlns:m="http://apache.org/dav/xmlns" xmlns:C="svn:"> <C:error/> <m:human-readable errcode="720003"> Could not open the requested SVN filesystem </m:human-readable> </D:error>

  Read the article

• How to force mdadm to stop RAID5 array?

  - by lucek

  I have /dev/md127 RAID5 array that consisted of four drives. I managed to hot remove them from the array and currently /dev/md127 does not have any drives: cat /proc/mdstat Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10] md0 : active raid1 sdd1[0] sda1[1] 304052032 blocks super 1.2 [2/2] [UU] md1 : active raid0 sda5[1] sdd5[0] 16770048 blocks super 1.2 512k chunks md127 : active raid5 super 1.2 level 5, 512k chunk, algorithm 2 [4/0] [____] unused devices: <none> and mdadm --detail /dev/md127 /dev/md127: Version : 1.2 Creation Time : Thu Sep 6 10:39:57 2012 Raid Level : raid5 Array Size : 8790402048 (8383.18 GiB 9001.37 GB) Used Dev Size : 2930134016 (2794.39 GiB 3000.46 GB) Raid Devices : 4 Total Devices : 0 Persistence : Superblock is persistent Update Time : Fri Sep 7 17:19:47 2012 State : clean, FAILED Active Devices : 0 Working Devices : 0 Failed Devices : 0 Spare Devices : 0 Layout : left-symmetric Chunk Size : 512K Number Major Minor RaidDevice State 0 0 0 0 removed 1 0 0 1 removed 2 0 0 2 removed 3 0 0 3 removed I've tried to do mdadm --stop /dev/md127 but: mdadm --stop /dev/md127 mdadm: Cannot get exclusive access to /dev/md127:Perhaps a running process, mounted filesystem or active volume group? I made sure that it's unmounted, umount -l /dev/md127 and confirmed that it indeed is unmounted: umount /dev/md127 umount: /dev/md127: not mounted I've tried to zero superblock of each drive and I get (for each drive): mdadm --zero-superblock /dev/sde1 mdadm: Unrecognised md component device - /dev/sde1 Here's output of lsof|grep md127: lsof|grep md127 md127_rai 276 root cwd DIR 9,0 4096 2 / md127_rai 276 root rtd DIR 9,0 4096 2 / md127_rai 276 root txt unknown /proc/276/exe What else can I do? LVM is not even installed so it can't be a factor.

  Read the article

• Need a place to store a few bytes of meta information on storage media

  - by Jason C

  I'm working on an embedded project. I need a place to store some filesystem-independent meta information on a storage device. The device has an MSDOS partition table. The device also may have unallocated space (depending on its size) but it will be TRIMmed (and also may be blown away by new partitions in the future). I need a location on the device that is not unallocated and that has a low risk of being touched (outside of completely erasing the device). The device is only guaranteed to have an MBR at the point the meta data needs to first be written; meaning there are no EBRs/VBRs present that I could use. There are 446 bytes at the very start of the device available for MBR bootstrap code. Currently my only idea is to store data at the end of this block. However, the device is bootable and I have no way of knowing if I'd be blowing away bootstrap code or not. The sector size is 512 bytes and the MBR is the first sector, I'm pretty sure (correct me if I'm wrong) that that means the second sector is available for use by partition data, so I can't use that either. Does anybody have any ideas? I need 4 bytes of space.

  Read the article

• Sparc Solaris 2.6 will not boot

  - by joshxdr

  I have a very old Sparc Solaris network that was working fine last week, but after a power outage none of the workstations will boot. The network looks like this: host A: solaris 2.6, shares /export/home to network by NFS host B: solaris 8, runs NIS server. Mounts /export/home/ by NFS. host C: RHEL5, shares /share to network by NFS. Mounts /export/home/ by NFS. I figured that the main problem was host A, since you need the home directories available for the other workstations to boot(?). Host A does not mount anything by NFS as far as I know. However, this workstation will NOT boot. The OBP bootup sequence looks like this: Boot device <blah> configuring network interface le0 Hostname <hostname> check file system <everything ok> check ufs filesystem <everything ok> NIS domainname is <name> starting router discovery starting rpc services: rpcbind keyserv ypbind done setting default interface for multicast: add net 224.0.0.0: gateway <hostname> <HANGS at this point> Is there some kind of debug mode so that I can get more detail as to why the workstation won't boot? Is my network structure inherently susceptible to power outage? Is there a way I can boot up to command line so I can at least turn off the NFS mounting?

  Read the article

• Ubuntu 9.04: Ripping CDs with grip?

  - by chris

  I tried to rip a CD tonight, and couldn't figure out how to configure grip - /dev/cdrom doesn't seem to be the mount point for music CDs any more. How can I configure grip to find CDs? Update: /etc/fstab has /dev/scd0 /media/cdrom0 udf,iso9660 user,noauto,exec,utf8 0 0 But there's nothing visible in /media/cdrom0 (or /media/cdrom, which is a symlink to cdrom0) There's an icon on the desktop labeled "Audio Disk" and opening it shows the .wav files on the CD. The location is cdda://sr0/, but grip doesn't like that either. Trying to manually mount /dev/sr0, I get $ sudo mount -t auto /dev/sr0 foo/ mount: block device /dev/sr0 is write-protected, mounting read-only mount: you must specify the filesystem type Update 2: Tried to change the media handling preferences (From a file browser, Edit-Preferences, Media, CD Audio) to "Do Nothing". CD Still doesn't mount. Update 3: With an audio CD in the drive: $ ls -l /dev/ | grep cd lrwxrwxrwx 1 root root 3 2009-09-15 22:13 cdrom1 -> sr0 lrwxrwxrwx 1 root root 3 2009-09-15 22:13 cdrw1 -> sr0 drwxr-xr-x 2 root root 60 2009-09-15 22:13 pktcdvd lrwxrwxrwx 1 root root 3 2009-09-15 22:13 scd0 -> sr0 crw-rw----+ 1 root cdrom 21, 2 2009-09-15 22:13 sg2 brw-rw----+ 1 root cdrom 11, 0 2009-09-15 22:13 sr0

  Read the article

• STOP 0x7b booting from iSCSI

  - by Michael

  Hi, I've a Windows 2008 SBS running. It boots of iSCSI. That setup worked for months until yesterday. I intended to reboot and gained a: STOP 0x0000007b INACCESSIBLE_BOOT_DEVICE and no idea why. My setup hasn't changed. No new controller, no new or changed iSCSI targets, no new Network Card or IP address changes. I had all Windows Updates on it. Last known good: same STOP. Allow unsigned drivers: same STOP. Safe mode (all variants): same STOP. Mount target from a client: works. Filesystem check fine. I booted of the SBS DVD but in computer repair options my target doesn't appear. When i choose setup the target appears. So, how can i diagnose what's going wrong? Any helpful tools? Any hints? Thanks in advance Michael

  Read the article

• Migrating to AWS Cloud with auto-scaling - where to put Redis and ElasticSearch?

  - by RobMasters

  I've been trying to research this topic but haven't found anywhere that recommends where to install services such as Redis and ElasticSearch when migrating to a cloud framework. I'm currently running a Symfony2 application on 2 static servers - one is running MySQL and the other is the public facing web server, which also has Redis and ElasticSearch running on it. Both of these servers are virtualised, but they're static in terms of not being able to replicate at present (various aspects are still dependent on the local filesystem). The goal is to migrate to AWS and use auto-scaling to be able to spin up and kill web servers as required, but I'm not clear on what I should put on each EC2 instance. Should they be single-responsibility only? i.e. Set up individual instances for the web server(s), Redis, and ElasticSearch and most likely an RDS instance for MySQL and only set up auto-scaling on the web server(s)? I don't foresee having to scale the ElasticSearch server anytime soon as it's only driving the search functionality, but it's possible that Redis may need to be replicated at some point - but should this be done manually? I'm not sure of how this could be done automatically as each instance needs to be configured to know about it's master/slave(s) as far as I know. I'd appreciate advice on this. One more quick question while I'm here - how would I be able to deploy code changes when there are X web servers currently active? I'm using a Capifony deployment script (Symfony2 version of Capistrano), which I think can handle multiple servers easily enough by specifying an array of :domain addresses...but how can should this be handled when the number of web servers can vary?

  Read the article

• Replace Linux Boot-Drive | ext3 to btrfs

  - by bardiir

  I've got a headless server running Debian Linux currently. Linux vault 3.2.0-3-686-pae #1 SMP Mon Jul 23 03:50:34 UTC 2012 i686 GNU/Linux The root filesystem is located on an ext3 partition on the main harddrive. My data is located on multiple harddrives that are bundled to a storage pool running with btrfs. UUID=072a7fce-bfea-46fa-923f-4fb0827ae428 / ext3 errors=remount-ro 0 1 UUID=b50965f1-a2e1-443f-876f-578b5f93cbf1 none swap sw 0 0 UUID=881e3ad9-31c4-4296-ae60-eae6c98ea45f none swap sw 0 0 UUID=30d8ae34-e2f0-44b4-bbcc-22d761a128f6 /data btrfs defaults,compress,autodefrag 0 0 What I'd like to do is to place / into the btrfs pool too. The ideal solution would provide the flexibility to boot from any disk in the system alike, so if the main drive fails I'd just need to swap another one into the main slot and it would be bootable like the main one. My main problem is, everything I do needs to result in a bootable system that is open to ssh logins via network as this server is 100% headless so there is no possibility to boot it from a live cd or anything like that. So I'd like to be extra sure everything works out fine :) How would I best go about this? Can anybody hint me to guides or whip something up for these tasks? Anything I forgot to think about? Copy root-data into btrfs pool, adjust mountpoints,... Adjust GRUB to boot from btrfs pool UUID or the local device where GRUB is installed Sync GRUB to all harddrives so every drive is equally bootable (is this even possible without destroying the btrfs partitions on the drives or would I need to disconnect the drives, install grub on them and then connect them back with a slightly smaller partition?)

  Read the article

• Moving a site from IIs6 to IIS7.5

  - by Sukotto

  I need to move a site off of IIS6 (Win Server 2003) and onto IIS7.5 (Win Server 2008) as soon as possible. Preferably tomorrow. The site itself is a delightful mix of classic asp (vbscript) and one-off asp.net (C#) applications (each asp.net app is in its own virtual dir and has a self-contained web.config). In case it's relevant, this is a sort of research site made up of 40 or 50 unconnected microsites. Each microsite is typically a simple form allowing a user to submit a form, which then runs a Stored Proc on a sqlserver db and displays a chart and/or table of the results. There is very little security to worry about. The database connection info is in a central file (in the case of the classic asp) or app's individual web.config (lots of duplication there) To add a little spice to the exercise... I have no idea how to admin IIS The company no longer employs the sysadmin or the guys who set this thing up. (They're not going to employ me much longer either but my sense of professional pride does not permit me to just walk away from this task). The servers are on mutually firewalled networks and I have to perform a convoluted, multi-step process to copy anything from one to the other. Would someone please point me to a crash-course tutorial for accomplishing the above? I have: a complete copy of the site's filesystem on the new box installed the 3rd party charting tool on the new system a config.xml file from the "all tasks - save configuration to a file" right click menu. There doesn't seem to be a way to import it on the new system however. The newer IIS manager has a completely different UI and I'm totally lost. Please help.

  Read the article

• How can I use wildcards in an Nginx map directive?

  - by Ian Clelland

  I am trying to use Nginx to served cached files produced by a web application, and have spotted a potential problem; that the url-space is wide, and will exceed the Ext3 limit of 32000 subdirectories. I would like to break up the subdirectories, making, say, a two-level filesystem cache. So, where I am currently caching a file at /var/cache/www/arbitrary_directory_name/index.html I would store that instead at something like /var/cache/www/a/r/arbitrary_directory_name/index.html My trouble is that I can't get try_files, or even rewrite to make that mapping. My searching on the subject leads me to believe that I need to do something like this (heavily abbreviated): http { map $request_uri $prefix { /aa* a/a; /ab* a/b; /ac* a/c; ... /zz* z/z; } location / { try_files /var/cache/www/$prefix/$request_uri/index.html @fallback; # or # if (-f /var/cache/www/$prefix/$request_uri/index.html) { # rewrite ^(.*)$ /var/cache/www/$prefix/$1/index.html; # } } } But I can't get the /aa* pattern to match the incoming uri. Without the *, it will match an exact uri, but I can't get it to match just the first two characters. The Nginx documentation suggests that wildcards should be allowed, but I can't see a way to get them to work. Is there a way to do this? Am I missing something simple? Or am I going about this the wrong way?

  Read the article

• Borked ubuntu uninstall - need to delete boot partition (i think)

  - by Max Williams

  I just got a new pc laptop with windows 7 and wanted to install Ubuntu on it. Which i did, no problem there, by downloading the installer, burning it to dvd then booting off the dvd and installing. Then, i realised that the new Ubuntu 12.04 uses the Unity desktop, which i immediately disliked, and after some research, began to hate. So, i decided (after a little googling) to install Linux Mint instead. So, thinking i'd better start from scratch, i went to the Windows 7 disk manager and wiped the Ubuntu partition that had been created. Now, when i start up, i get an error from grub, the ubuntu boot manager: error: unknown filesystem grub rescue> _ and a blinking cursor where i can enter commands. I suspect that what i've done is deleted the main ubuntu partition but NOT deleted another partition which is a boot partition, or something like that? Can anyone tell me how i can rescue or unbork this? I'd like to either a) get back to my original windows-only setup OR b) install linux mint off dvd (which i have), into the empty partition, fixing any grub confusion in the process. Any suggestions? Thanks, max BTW please don't answer if you're just going to tell me to stick with 12.04, or install a different distro or something. I definitely want Mint and just want to fix this mess - thanks :)

  Read the article

• Synchronize two directories on linux pc

  - by Gab

  I need a distributed filesystem (or a synchronization tool) that is capable of keeping a directory synchronized across 4 pc. My requirements are: offline access (data must be available offline on each pc) preserve execution rights: some files are marked executable on a linux partition. This flag should be replicated. efficient sync strategy: some of my files are 20GB, they are changed quite often, but in very little parts (Virtualbox images). Delta transmissions are welcome. efficient handling of space: no history for files, files shouldn't be copied to temp directories "just in case you break it". it must propagate deletions of files modification can happen in any of the 4 pcs, they should be propagated when other pc are connected. Other specs of my solution are: Sync is over a lan, the total amount of data to be synced is around 180GB, in some ten thousand files. Changes are small, but can happen in big files. At the moment i'm interested in a linux only solution. conflicts either don't happen or are solved with "last one wins" I haven't found any good solution. I've been trying: unison: it is the only one working at the moment, but during the hashing phase it hangs my pc for some minute, disk light steady on. Sparkleshare doesn't handle large files nicely. It keeps an history of all your changes that grows up indefinitely. They promise it will be fixed in next releases, but at the moment it still doesn't fit my needs. owncloud (keeps history of each file i change) coda ? (help! i couldn't set it up correctly!) git-annex assistant transforms all your files in symlinks and mark the original file as read only ("just in case you make a mistake while you modify it"!). Before you edit a file you have to issue a special command "git-annex unlock", that creates a local copy of the file, and you have to remember to lock it again if you want it synchronized. What to try next?

  Read the article

• Caching/preloading files on Linux into RAM

  - by Andrioid

  I have a rather old server that has 4GB of RAM and it is pretty much serving the same files all day, but it is doing so from the hard drive while 3GBs of RAM are "free". Anyone who has ever tried running a ram-drive can witness that It's awesome in terms of speed. The memory usage of this system is usually never higher than 1GB/4GB so I want to know if there is a way to use that extra memory for something good. Is it possible to tell the filesystem to always serve certain files out of RAM? Are there any other methods I can use to improve file reading capabilities by use of RAM? More specifically, I am not looking for a 'hack' here. I want file system calls to serve the files from RAM without needing to create a ram-drive and copy the files there manually. Or at least a script that does this for me. Possible applications here are: Web servers with static files that get read alot Application servers with large libraries Desktop computers with too much RAM Any ideas? Edit: Found this very informative: The Linux Page Cache and pdflush As Zan pointed out, the memory isn't actually free. What I mean is that it's not being used by applications and I want to control what should be cached in memory.

  Read the article

• System occasionally hangs boot process with SLES 11

  - by ThaMe90

  I have several (new) systems on which I had to install SLES11 on. However, after a few (though not every) reboots, the system hangs during the boot sequence. It will only continue after I physically press a key on the keyboard. From what I've found in the dmesg log from a failed boot is the following: [ 22.170276] sd 0:0:0:0: [sda] Mode Sense: b7 00 00 08 [ 22.171155] sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA [ 22.182760] sda: sda1 sda2 sda3 [ 22.383424] sd 0:0:0:0: [sda] Attached SCSI disk [ 22.545372] PM: Marking nosave pages: 000000000009a000 - 0000000000100000 [ 22.545377] PM: Marking nosave pages: 00000000bf780000 - 0000000100000000 [ 22.546217] PM: Basic memory bitmaps created [ 22.590380] PM: Basic memory bitmaps freed [ 22.596284] PM: Starting manual resume from disk [ 22.602319] PM: Resume from partition 8:1 [ 22.602321] PM: Checking hibernation image. [ 22.602479] PM: Error -22 checking image file [ 22.602481] PM: Resume from disk failed. [ 22.718727] kjournald starting. Commit interval 15 seconds [ 22.718960] EXT3-fs (sda3): using internal journal [ 22.718964] EXT3-fs (sda3): mounted filesystem with ordered data mode [ 1555.644404] udevd version 128 started [ 1555.697664] input: Power Button as /devices/LNXSYSTM:00/LNXSYBUS:00/PNP0C0C:00/input/input0 [ 1555.707961] ACPI: Power Button [PWRB] I've looked around the internet for the PM: Resume from disk failed. message, but this seems to only be important when restoring the system after a hybernate, i.e. restore from the hdd. But this is not my situation. I only get this after a reboot, as I said before. The timestamp [ 1555.xxxxxx] is only the result of me pressing a key on the keyboard. Any suggestions on how to proceed? As I am getting stuck on this issue.

  Read the article

• Data recovery from corrupt Ubuntu partition/directory (question about a previous answer)

  - by JoshMaurice

  I have an Ubuntu installation that won't boot anymore. I asked my previous question about it here: http://superuser.com/questions/15916/ubuntu-chkdsk-equivalent Bolotov replied: As I see from your previous question you can boot Windows so you could use dskprobe from Windows XP Service Pack 2 Support Tools to make sure that fs type is correct ... but it's already correct fs type 7 is NTFS. Message "The type of the filesystem is RAW. CHKDSK is not available for RAW drives." means that windows can't determine fs type for some reason. As we see fs type is correct. To run Chkdsk on your Windows partition you can install Windows Recovery Console, boot in recovery console and check your disk. After checking the disk you will gain access to you c:\ubuntu\disks. I think you can mount your linux partition (which is in file) as usual loop-back device: mount -o loop [path to your linux-loopback-partition] But you should mount windows patrition first. So now I'd like to know: Within the recovery console I will be issuing the commands "chkdsk -r" and then "mount -o loop [path to windows partition]" and then "mount -o loop c:\ubuntu\disks", correct? I do have a ("corrupt and unreadable") c:\ubuntu\disks directory so that appears to be the correct path to the linux partition; do you know the path to the windows partition? would that be just "c:\"?

  Read the article

• Virtualbox Networking: XP Guest, Ubuntu Host: Connecting to Windows servers & local network?

  - by user51833

  Here's what I have: Windows XP running in VirtualBox 3.0.8_OSE r53138; Host OS = Ubuntu 9.10 "Karmic Koala"; Windows network in my office with smb fileservers; Guest OS is connected to the internet and is sharing folders with Host OS; Limited networking expertise. Here's what I actually need to do: Use MS Outlook in my XP guest with all its calendar-sharing features and stuff (if this is all done through the internet then great) - or find a Linux app that can do the same stuff; Map Windows network servers, eg. smb://server01/ in my XP guest (I can already access these in Ubuntu. Here's what I've tried with no luck: Entering the server address (example above) in my XP guest windows explorer address bar (got a "could not access the file, path or drive" error message - maybe if I could enter login/pass information? But I don't know how); Mapping the server as a network drive (Windows could not find the path); Mounting the server as one of my shared folders (I couldn't find it through the shared folders browser in VirtualBox - is there somewhere in the Linux filesystem that Ubuntu keeps links to mounted servers?).

  Read the article

• Large, high performance object or key/value store for HTTP serving on Linux

  - by Tommy

  I have a service that serves images to end users at a very high rate using plain HTTP. The images vary between 4 and 64kbytes, and there are 1.300.000.000 of them in total. The dataset is about 30TiB in size and changes (new objects, updates, deletes) make out less than 1% of the requests. The number of requests pr. second vary from 240 to 9000 and is dispersed pretty much all over, with few objects being especially "hot". As of now, these images are files on a ext3 filesystem distributed read only across a large amount of mid range servers. This poses several problems: Using a fileysystem is very inefficient since the metadata size is large, the inode/dentry cache is volatile on linux and some daemons tend to stat()/readdir() it's way through the directory structure, which in my case becomes very expensive. Updating the dataset is very time consuming and requires remounting between set A and B. The only reasonable handling is operating on the block device for backup, copying, etc. What I would like is a deamon that: speaks HTTP (get, put, delete and perhaps update) stores data it in an efficient structure. The index should remain in memory, and considering the amount of objects, the overhead must be small. The software should be able to handle massive connections with slow (if any) time needed to ramp up. Index should be read in memory at startup. Statistics would be nice, but not mandatory. I have experimented a bit with riak, redis, mongodb, kyoto and varnish with persistent storage, but I haven't had the chance to dig in really deep yet.

  Read the article

• Ngins wont send POST to fastcgi backend, but GET works fine?

  - by xyld

  Not sure why, but it is happy sending a GET to the fastcgi backend (Mercurial hgwebdir in this case), but simply resorts to the filesystem if the request is a POST. Relevant parts of nginx.conf: location / { root /var/www/htdocs/; index index.html; autoindex on; } location /hg { fastcgi_pass unix:/var/run/hg-fastcgi.socket; include fastcgi_params; if ($request_uri ~ ^/hg([^?#]*)) { set $rewritten_uri $1; } limit_except GET { allow all; deny all; auth_basic "hg secured repos"; auth_basic_user_file /var/trac.htpasswd; } fastcgi_param SCRIPT_NAME "/hg"; fastcgi_param PATH_INFO $rewritten_uri; # for authentication fastcgi_param AUTH_USER $remote_user; fastcgi_param REMOTE_USER $remote_user; #fastcgi_pass_header Authorization; #fastcgi_intercept_errors on; } GET's work fine, but POST delivers this error to the error_log: 2010/05/17 14:12:27 [error] 18736#0: *1601 open() "/usr/html/hg/test" failed (2: No such file or directory), client: XX.XX.XX.XX, server: domain.com, request: "POST /hg/test HTTP/1.1", host: "domain.com" What could possibly be the issue? I'm trying to allow read-only access via GET's to the page, but require authorization when using hg push to the same url which sends a POST request.

  Read the article

• How do I prevent a tar pipe from causing swapping?

  - by Jeff Shattock

  I have a rather large filesystem that I need to transfer from one Linux server to another. I figured the best way to do this was via a tar/netcat pipe arrangment, something like tar c . | pv | nc blah blah blah And it works great, the network stays fairly saturated, life is good. Until the source machine starts swapping. The files are on a raid on the source system, so the read speed is much faster than the write speed on the other end. Since the dest machine hasnt picked up the data yet, the source machine needs to stick it somewhere, so into RAM it goes, until there is no more free RAM. It then starts swapping, which is horribly painful since that machine has its OS installed on a somewhat slow CF card. Both machines have 4GB of physical ram, 64 bit Ubuntu 9.04 server. GigE link between them. How do I prevent this swapping? Can I put a "speed-limit" on the tar or netcat process so that the transfer speed doesn't overwhelm the write throughput on the destination end? The man pages didn't list anything, but there might be something I'm overlooking.

  Read the article

< Previous Page | 73 74 75 76 77 78 79 80 81 82 83 84  | Next Page >