unix linux and storageunix linux and storageUnix Linux Storage , netapp, performance tuning, automation, hacks, tips, documents and useful hints Articles
How to resize Netapp Qtree Quotas .. ?
2008-05-28 08:01:00 1) Get the quotas file from netapp [root@unixfoo-l-adm root]# rsh unixfoo-nfs1 rdfile /etc/quotas > /tmp/quotas2) Edit the quotas file and change the quota for the required qtree. [root@unixfoo-l-adm root]# vi /tmp/quotas Example : Edit quotas as below /vol/vol1/qtree1 tree 250G #(Old entry) to /vol/vol1/qtree1 tree 300G #(New entry) 3) Upload the quotas file to Netapp [root@unixfoo-l-adm root]# rsh unixfoo-nfs1 wrfile /etc/quotas < /tmp/quotas4) Use Quota resize command to resize the qtree quota. [root@unixfoo-l-adm root]# rsh unixfoo-nfs1 quota resize vol15) Done. More About: Resize
Useful AIX tips
2008-05-08 16:18:00 Display paging space details# lsps -aPage Space Physical Volume Volume Group Size %Used Active Auto Typehd6 &nbs p; hdisk1 ; rootvg ; 8192MB 1 yes yes lv#Memory module details : # lscfg -vp | egrep "Memory DIMM|Size" Memory DIMM: Size........................1024 &nb sp; Memory DIMM: Size........................1024 &nb sp; Memory DIMM: Size........................1024 &nb sp; Memory DIMM: Size........................1024&nb... More About: Tips
Linux RPM database recovery
2008-05-07 13:29:00 Sometimes you may get the below error while running "rpm -qa" or any other rpm listing command. This error suggests that the RPM db is corrupted and you need to clean it up.rpmdb: PANIC: fatal region error detected; run recoveryerror: db4 error(-30977) from dbenv->open: DB_RUNRECOVERYHere are the steps to do it.Remove the rpm db lock information filesrm -f /var/lib/rpm/__db*Run the rpm rebuild database command . The rpm --rebuilddb option will rebuild the database indices from the installed RPM package headers. This process will take a while to complete. Once it is completed, the rpm -qa and other rpm listing commands will work.rpm --rebuilddb More About: Linux , Recovery , Database
Increasing swap performance
2008-05-06 10:46:00 If your Linux machine is configured with several swap partitions, the default trend is to use of one partition at a time. Once one swap partition has been full the next will be used. This is not always the best method for performance, since when a new process needs to be swapped to disk it may be forced to wait until another process is swapped out. For instance, if there are two swap partitions specified in /etc/fstab it will look something like this: /dev/sda2 swap swap defaults 0 0/dev/sda3 swap swap defaults 0 0Change the mount options section from ''defaults'' to ''pri=0'' : /dev/sda2 swap swap pri=0 0 0/dev/sda3 swap swap pri=0 0 0If you want to do this in a live system, then swapoff and swapon with "-p 0" option for each swap device - one by one. Once this has been done the system will be able to access any of the designated swap partitions independently of the others. This can increase ... More About: Performance , Swap
Netapp deduplication
2008-04-23 14:37:00 Netapp deduplication - is a new technology to control the rate of data growth. The average UNIX or Windows disk volume contains thousands or even millions of duplicate data objects. As data is created, distributed, backed up, and archived, duplicate data objects are stored unabated across all storage tiers. The end result is inefficient utilization of data storage resources. By eliminating redundant data objects and referencing just the original object, huge benefit is obtained through storage space efficiencies - which results in "Cost benefit" and "Management benefit".Deduplication implementation guide : http://media.netapp.com/documents/tr-3505 .pdfDeduplication estimation calculator : http://www.dedupecalc.com/
HPUX tips
2008-04-22 13:14:00 1) Find number of processors in the machine.# /usr/sbin/ioscan -k -C processorH/W Path Class Description============================== =160 processor Processor166 ; processor Processor#2) Find the model of the hardware# model9000/800/L2000-44#3) Find the network devices attached to the server# /usr/sbin/ioscan -k -C lanH/W Path Class Description============================== =========0/0/0/0 & nbsp; lan HP PCI 10/100Base-TX Core0/2/0/0 ; lan HP A5230A/B5509BA PCI 10/100Base-TX Addon#4) List the device drivers in the system.# lsdev Character Block Driver&n... More About: Tips
Solaris LDOM virtualization documents
2008-04-18 12:56:00 Solaris LDOM virtualization document links : Overview : http://www.sun.com/servers/coolthreads/ld oms/wp.pdfLDOMs Beginners guide : http://www.sun.com/blueprints/0207/820-08 32.pdfLDOMs presentation : http://www.snpnet.com/customer_pub/sun/is v_LDOM/Admin guide : http://docs-pdf.sun.com/819-6428-11/819-6 428-11.pdf More About: Virtualization , Solaris , Documents
AIX performance monitoring : filemon
2008-04-11 17:29:00 "filemon" command monitors the performance of the file system, and reports the I/O activity. The I/O activity monitored is based on the time interval during which the filemon is run. It reports on various levels of file system utilization, including the LVM, virtual memory, and physical disk layers. Without any options, it runs in the background while application programs or system commands are being run and monitored. The trace starts automatically until it is stopped. At that time, the command generates an I/O activity report and exits. "trcstop" command is used to stop the filemon performance capturing.A sample run belowunixfoo-aix:# filemonRun trcstop command to signal end of trace.unixfoo-aix:# Fri Apr 11 08:17:42 2008System: AIX 5.3 Node: unixfoo-aix Machine: XXXXXXXXXXXunixfoo-aix:#unixfoo-aix:# trcstopFri Apr 11 08:19:30 2008System: AIX 5.3 Node: unixfoo-aix Machine: XXXXXXXXXXXCpu utilization: 57.9%Cpu allocation: 91.2%58051805 events were lost. Repo... More About: Performance , Monitoring
HP-UX : Kernel Parameter Recommendations
2008-04-10 16:03:00 Below is the list of commonly recommended HPUX kernel parameters. But this may vary based on your application need.# Parameter Value bufpages 0 # on HP-UX 10.X create_fastlinks 1 dbc_max_pct 25 fs_async 1 maxdsiz 2063806464 maxfiles 200 maxfiles_lim 2048 maxssiz (80*1024*1024) maxswapchunks 4096 maxtsiz (1024*1024*1024) maxuprc 200 maxusers 124 netmemmax 0 # on desktop systems -1 # on data servers nfile 2800 ninode 15000 # 4000 on HP-UX 10.20 multi-processor systems nproc 1024 npty 512More details at : http://docs.hp.com/en/1219/tuningwp.html# kernparmrec More About: Kernel
HPUX : Identify 32-bit or 64-bit kernel
2008-04-10 15:09:00 Here are the steps to identify whether you are running a 32-bit or 64-bit kernel.getconf KERNEL_BITS will return whether it is a 32 bit or 64 bit. hpux1 # getconf KERNEL_BITS 32 &nb sp; &n bsp; & nbsp; ; --> 32-bit hpux2 # getconf KERNEL_BITS 64 &nb sp; &n bsp; & nbsp; ; --> 64-bit Check the vmunix's filetype. hpux1 # file /stand/vmunix /sta... More About: Kernel
Oracle's Direct NFS
2008-04-05 17:32:00 Direct NFS (DNFS) is an Oracle Database 11g implementation of NFS client that runs as part of the Oracle Database 11g engine. Through this integration, the Oracle database engine optimizes the I/O access path to NFS server to provide improved scalability and reliability. In previous Oracle versions, the I/O operations to the NFS server were done through the native NFS client of the Operating system. Even though the native NFS client implementations evolved over the last few years, the features & configuration varied across operating systems and the end user experience was not uniform across them. Each operating system needed a special set of NFS mount options to maximize its performance.Oracle Direct NFS client provides uniform features across all operating systems, independent of native NFS client implementation. Manual tuning of NFS client parameters and mount options are no longer required to maximize the performance. Direct NFS client is available across all operating system... More About: Direct
Few /proc files
2008-04-04 18:21:00 /proc/cmdline - file has the parameters passed to the kernel while the system boots-up. /proc/kcore - This file represents the physical memory of the system and is stored in the core file format. The size of this file is equal to the size of the physical memory (RAM) used plus 4 KB./proc/filesystems - This file displays a list of the file system types currently supported by the kernel. /proc/kmsg - This file is used to hold messages generated by the kernel. These messages are then picked up by other programs, such as /sbin/klogd or /bin/dmesg./proc/stat - This file keeps track of a variety of different statistics about the system since it was last restarted. More About: Proc , Files
Know about Netapp Snapvault
2008-04-04 17:20:00 Netapp SnapVault is a heterogeneous disk-to-disk backup solution for Netapp filers and systems with other OS (Solaris, HP-UX, AIX, Windows, and Linux). In event of data loss or corruption on a filer, backed up data can be restored from the SnapVault secondary storage system with less downtime and less of the uncertainty associated with conventional tape backup and restore operations.About Snapvault : http://www.netapp.com/us/products/protect ion-software/snapvault.htmlOpen systems snapvault : http://www.netapp.com/us/products/protect ion-software/ossv.htmlPresentation on Snapvault features : http://www-download.netapp.com/edm/TT/WOD /WOD20070222/index.html
AIX : Managing service subsystem
2008-03-29 06:05:00 lssrv , stopsrc and startsrc are used to manage service subsystem in AIX. lssrc lists out all services that are active and inactive. "startsrc" and "stopsrc" are used to start and stop subsystems.# lssrc -aSubsystem Group PID &n bsp; Status syslogd &nbs p; ras &n bsp; 200938 ; active inetd ; tcpip 213152 ; active portmap &nbs p; portmap &nbs p; 135278 ; active xntpd &nb... More About: Service
Netapp VIFs loadbalancing
2008-03-18 17:14:00 NetApp loadbalancing algorithm for determining ethernet load balancing (vifs) differs from Cisco, so it's normal to see unbalanced traffic in Netapp vifs. The Cisco method for determining load balancing is : http://www.cisco.com/warp/public/473/4.ht ml The NetApp method is described as:((source_address XOR destination_address) % number_of_links) -- where source and destination take the last byte of the IP or MAC address to perform the calculation.Example : Filer IP is 10.128.60.1clients are 10.128.60.[3-6](1 xor 3) % 3 == 2 % 3 == link#2 (e0d)(1 xor 4) % 3 == 5 % 3 == link#2 (e0d)(1 xor 5) % 3 == 4 % 3 == link#1 (e0c)(1 xor 6) % 3 == 7 % 3 == link#1 (e0c) If a client used 10.128.60.2 or 10.128.60.9, then we would see outgoing traffic on e0b. MAC address hashing calculation it would be more imbalanced than the IP address method:(0xB6 xor 0x4E) % 3 == 0xF8 % 3 == 248 % 3 == link#2 (e0d)(0xB6 xor 0xD8) % 3 == 0x6E % 3 == 110 % 3 == link#2 (e0d)(0xB6 xor 0x56...
Steps to create initrd
2008-03-06 18:31:00 Creating initrd using mkinitrd# mkinitrd -v -f --with=e1000 /tmp/initrd.gz 2.6.9-55Extracting the initrd image to add contents in it# mkdir /initrdroot# cd /initrdroot# gzip -dc /tmp/initrd.gz | cpio -id# ls /initrdrootbin dev etc init lib linuxrc lost+found mnt proc root sbin sys sysroot tmp usr var#Now add binaries and libraries to the initrdrootTo recreate the initrd based on /initrdroot# find ./ | cpio -H newc -o > /tmp/unixfooinitrd# gzip /tmp/unixfooinitrd# mv /tmp/unixfooinitrd /boot/unixfooinitrd.gz More About: Create , Steps
Resizing ext3 filesystem
2008-03-03 12:27:00 The ext3 filesystem on Linux can be resized using ext2online . This can only be done when the filesystem is online (mounted) and where the the filesystem is on a resizable logical volume. More details at http://ext2resize.sourceforge.net/Note : I have'nt tried this program and there is no warranty for this program. Use it carefully. More About: Filesystem
Windows provisioning from PXE
2008-03-03 11:27:00 Two opensource solutions through which you can provision Windows servers by network booting from a Linux PXE : http://unattended.sourceforge.net/http:// unattended-gui.sourceforge.net/Good one & they worked for me.
NFS mount options
2008-02-21 10:23:00 Some important nfs mount options in Linux. tcp - Specifies for the NFS mount to use the TCP protocol instead of UDP.rsize=8192 and wsize=8192 - These settings speed up NFS communication for reads (rsize) and writes (wsize) by setting a larger data block size, in bytes, to be transferred at one time. Do performance tests before changing these values.hard or soft - Specifies whether the program using a file via an NFS connection should stop and wait (hard) for the server to come back online if the host serving the exported file system is unavailable, or if it should report an error (soft). If hard is specified, the user cannot terminate the process waiting for the NFS communication to resume unless the intrsoft, is specified, the user can set an additional timeo=<value> option, where <value> specifies the number of seconds to pass before the error is reported. option is also specified.nolock - Disables file locking. This setting is occasionally required when connect... More About: Options , Mount
Netapp Flexshare
2008-02-18 15:59:00 A very good demo on Netapp Flexshare and how it impacts performance : http://www.netapp.com/ftp/flash/flexshare -demo/flexshare_demo.html
Know about /proc/meminfo
2008-02-14 08:35:00 [root@unixfoo root]# cat /proc/meminfo &nbs p; total: used: free: shared: buffers: cached:Mem: 8298250240 2093281280 6204968960 & nbsp; 0 177729536 1242714112Swap: 17174069248 0 17174069248MemTotal: &nb sp; 8103760 kBMemFree: & nbsp; 6059540 kBMemShared: ; 0 kBBuffers: & nbsp; 173564 kBCached: &n bsp; 1213588 kBSwapCached: &nbs p; 0 kBActive: &n bsp; 1062128 kBActiveAnon: 173804 kBActiveCache: 888324 kBInact_dirty: 358336 kBInact_laundry: 140060 kBInact_clean: &nb sp; ... More About: Proc
Converting ext2 filesystem to ext3 filesystem
2008-02-13 10:31:00 ext3 filesytem is a journal'd filesystem. The command "tune2fs" can add a journal to an existing ext2 file system and convert it to ext3, without altering the data already on the partition. To convert an ext2 file system to ext3, log in as root and type: /sbin/tune2fs -j /dev/hdbX In the above command, replace /dev/hdb with the device name and X with the partition number. More About: Filesystem
Xen 4.1 beta released
2008-02-05 04:09:00 Citrix XenServer 4.1 beta is released, with below features and improvementsScalability and Performance Increased number of simultaneous running VMsEnhanced nested page table (NPT) support for modern AMD processorsVLAN support in Standard EditionImproved Citrix Presentation Server performance and maximum number of user sessions Reliability and Manageability Host NIC bonding for fail-over (configured via CLI)Centralized loggingConfiguration of network management interfaces via the CLIUpdate/patch management integrated in XenCenterJava bindings for XenAPI in SDK Storage Initial shared fibre channel storage support (via CLI only)Enhanced support for NetApp filers, including snapshot and cloningWindows guest Hot disk removeISCSI improvementsSupport for hot-plugging USB storage as a storage repository Host System Rolling pool upgrade supportNIC driver updates (e1000, BNX2, TG3)Support several 10Gb network adapters (Mellanox/Chelsio)Improved hardware support Guest Support ... More About: Beta , Released
Network bonding : Part II ( modes of bonding )
2008-02-01 09:59:00 The steps for creating network boding in Linux is available in http://unixfoo.blogspot.com/search/label/ networking. RHEL bonding supports 7 possible "modes" for bonded interfaces. These modes determine the way in which traffic sent out of the bonded interface is actually dispersed over the real interfaces. Modes 0, 1, and 2 are by far the most commonly used among them.Mode 0 (balance-rr)This mode transmits packets in a sequential order from the first available slave through the last. If two real interfaces are slaves in the bond and two packets arrive destined out of the bonded interface the first will be transmitted on the first slave and the second frame will be transmitted on the second slave. The third packet will be sent on the first and so on. This provides load balancing and fault tolerance.Mode 1 (active-backup)This mode places one of the interfaces into a backup state and will only make it active if the link is lost by the active interface. Only one slave in the bond is ac... More About: Network , Part
How to configure Linux machine as router?
2008-02-01 09:26:00 Here are some simple steps to do it, using Network Address Translation (NAT)Enable packet forwarding . Make it permanent by adding "net.ipv4.ip_forward = 1" to /etc/sysctl.confecho "1" > /proc/sys/net/ipv4/ip_forwardEnable iptables to handle NAT. ( eth0 is the public connection )/sbin/iptables -t nat -A POSTROUTING -o eth0 -j MASQUERADESave iptables settings service iptables saveDone. Check by routing table by verifying "netstat -nr" and iptables rules by "iptables -L"This is a simple method for Redhat Enterprise Linux . See iptables docs for more detailed info. More About: Machine , Router , Configure
AIX tips : Switching to 64-bit kernel
2008-01-31 17:01:00 The AIX kernel is available in 32-bit or 64-bit mode. If a 32-bit kernel is running and you want to switch to 64-bit kernel , use the following command. ( provided the system has the “bos.mp64” file set installed. )# ln –sf /usr/lib/boot/unix_64 /unix# ln –sf /usr/lib/boot/unix_64 /usr/lib/boot/unix# bosboot –ak /usr/lib/boot/unix_64# shutdown –Fr More About: Tips , Switching , Kernel , Witch
Sticky bit
2008-01-30 17:23:00 What is the "sticky bit"? (t and T)The sticky bit is primarily used on shared directories. It is useful for shared directories such as /var/tmp and /tmp/ because users CAN CREATE files, READ and EXECUTE files owned by other users, but ARE NOT ALLOWED to REMOVE files owned by other users. Of course, users CAN REMOVE their own files. The sticky bit (file mode bit 01000) is used to indicate special treatment of certain files and directories. A directory for which the sticky bit is set restricts deletion of files it contains. A file in a sticky directory may only be removed or renamed by a user who has write permission on the directory, and either owns the file, owns the directory, or is the super-user. This is useful for directories such as /tmp, which must be publicly writable, but should deny users permission to arbitrarily delete or rename the files of others.Identifying sticky permissions.# ls -ld /var/tmpdrwxrwxrwt 2 ...
Linux Benchmarking tools
2008-01-30 10:21:00 Here is a list of few benchmarking tools put togetherBonnie++ : This is a benchmark suite that is aimed at performing a number of simple tests of hard drive and file system performance.Iozone : Iozone is a filesystem benchmark tool. The benchmark generates and measures a variety of file operations. Iozone has been ported to many machines and runs under many operating systems. Bonnie : This is a I/O throughput benchmarking toollmbench : This is a simple and portable benchmarking tool, which analyzes memory, cpu of a system.Netperf : Netperf is a benchmark that can be used to measure the performance of many different types of networking. It provides tests for both unidirecitonal throughput, and end-to-end latency.Cachebench : Benchmarking tool for memory subsystemSIO : Netapp performance analysis and benchmarking tool. More About: Linux , Tools
nmon : AIX and Linux performance monitoring tool
2008-01-29 10:46:00 Nmon : A free performance monitoring script for Linux and AIX systems. It gives lot of performance information and can output the performance analysis results in many format. nmon Document and download details @ http://www-941.haw.ibm.com/collaboration/ wiki/display/WikiPtype/nmon More About: Performance , Tool , Monitoring
What is Direct I/O
More articles from this author:2008-01-29 10:29:00 A file is simply a collection of data stored on media. When a process wants to access data from a file, the operating system brings the data into main memory, the process reads it, alters it and stores to the disk . The operating system could read and write data directly to and from the disk for each request, but the response time and throughput would be poor due to slow disk access times. The operating system therefore attempts to minimize the frequency of disk accesses by buffering data in main memory, within a structure called the file buffer cache.Certain applications derive no benefit from the file buffer cache. Databases normally manage data caching at the application level, so they do not need the file system to implement this service for them. The use of a file buffer cache results in undesirable overheads in such cases, since data is first moved from the disk to the file buffer cache and from there to the application buffer. This “doublecopying” of data results in more ... More About: Direct , Rect 1, 2, 3 |
|
 |
|
 |
