Unix Profession Webcast October 2007 - Circling Cycle

Unix Profession Webcast October 2007 Circling Cycle-ppt Download

  • Date:30 Jun 2020
  • Views:29
  • Downloads:0
  • Size:534.37 KB

Share Presentation : Unix Profession Webcast October 2007 Circling Cycle

Download and Preview : Unix Profession Webcast October 2007 Circling Cycle

Report CopyRight/DMCA Form For : Unix Profession Webcast October 2007 Circling Cycle


Transcription:

HP UX Swap and DumpBy Unix Linux Apprentice with 26Years of ExperienceDusan BaljevicSydney Australia.
Why This Document Frequent abuse of good design principles A friend in need is a friend indeed whystandard swap dump design fails in real scenarios Everyone has different opinion why not help.
system administrators and architects stopimplementing bad practices Especially important on large RAM servers Based on 26 year practical experiences inUnix Linux .
This Document is Not A replacement for HP s official statements A written manual to learn HP UX and itsdesign principles in detail Glorified personal experience to prove that.
I know best rather the opposite HP UX Current OfficialRecommendations Part 1Use the following guidelines when configuring swap Interleave device swap areas for better.
performance Two swap areas on different disks perform betterthan one swap area with the equivalent amount ofspace This configuration allows interleavedswapping which means the swap areas are written.
to concurrently thus enhancing performance When using LVM set up secondary swap areaswithin logical volumes that are on different disksusing lvextend If you have only one disk and must increase swap.
HP UX Current OfficialRecommendations Part 2 Similar sized device swap areas work best Deviceswap areas must have similar sizes for bestperformance Otherwise when all space in the.
smaller device swap area is used only the largerswap area is available making interleavingimpossible By default primary swap is located on the samedisk as the root file system The kernel.
configuration file contains the configurationinformation for primary swap If you are using logical volumes as secondary swap allocate the secondary swap to reside on a diskother than the root disk for better performance .
Disable mirror consistency checking for swapmirrored primary swap device no need to recover HP UX How Much Swap is Enough Every admin and architect has a different opinion Traditional views typically use formula .
SWAP 1 or 2 x RAM Some old designs and applications required even 3x RAM or more Old HP UX releases had serious issue with nowobsolete kernel parameter swapmem on see next.
HP UX How Much Dump is Enough The vast majority of problems are found in the kernel area rarely do the program data areas need to be examined rarely the shared memory areas and virtually never thebuffer file.
cache and shared libraries If a full crash dump is taken the total space needed withhigh as RAM and a bit more By compressed dump overall time taken will be reduced by HP UX How Much Dump is Enough .
crashconf vCrash dump configuration has been changed since boot CLASS PAGES INCLUDED IN DUMP DESCRIPTION UNUSED 9572004 no by default unused pages.
USERPG 1341553 no by default user process pagesBCACHE 1980 no by default buffer cache pagesKCODE 9142 no by default kernel code pagesUSTACK 1567 yes by default user process stacksFSDATA 12 yes by default file system metadata.
KDDATA 1492949 yes by default kernel dynamic dataKSDATA 8816 yes by default kernel static dataSUPERPG 128677 no by default unused kernel super pagesTotal pages on system 12556700Total pages included in dump 1503344.
Dump compressed ONDump Parallel ONDEVICE OFFSET kB SIZE kB LOGICAL VOL NAME 1 0x000005 2349920 4194304 64 0x000002 dev vg00 lvol2.
getconf PAGESIZE HP UX PseudoswapPseudoswap allows the kernel to treat a portion of physical memory as if it isswap space in order to satisfy the swap reservation policy Pseudo swap is.
enabled by default in all current versions of HP UX and is removed as kernelparameter in11i v3 swapmem on I have 2GB of swap With Pseudoswap swapmem on 1 and 8GB of availablememory Can I start a.
2GB Device Swap4GB process on 6GB Pseudo Swap 75 an idle server 8GB 8GB Reservable SwapWithout Pseudoswap swapmem on 0 .
No 2GB Device Swap 0GB Pseudo Swap2GB Reservable Swap Example of an Application SwapRequirements.
Please see SAP note 1112627 for a detailedexplanation of swap sizing and pseudo swap In general device swap configurations of 1 5 or 2 xRAM have proven appropriate for the majority ofSAP installations The recommendation is to set.
device swap to 2 x RAM minimum 20 GB Please refer to SAP note 153641 for a detailedexplanation of swap requirements on a per SAPinstance basis Basics of Crash Dumps.
Bad Example of Swap Design usr sbin swapinfo tmMb Mb Mb PCT START MbTYPE AVAIL USED FREE USED LIMIT RESERVE PRI NAMEdev 30464 0 30464 0 0 1 dev vg00 lvol2.
dev 30464 0 30464 0 0 1 dev vg00 swap1dev 30464 0 30464 0 0 1 dev vg00 swap2dev 30464 0 30464 0 0 1 dev vg00 swap3reserve 46202 46202memory 98292 2278 96014 2 .
total 220148 48480 171668 22 0 HP UX Maximum Swap Swap space in the kernel is managed using chunks of physical device space Thesecontain one or more usually more pages of.
memory but provide another layer of similar to inodes in file systems to keepthe globalswap table relatively small as opposed to atable indexed by swap page .
swchunk controls the size in physical disk Maximum Swap on HP UX Before 11i The total bytes of swap space manageable bysystem on HP UX 11i older releases is swchunk x 1KB x 16384.
where16384 is the system maximum number ofswap chunks in the swap table as defined bykernel parameter maxswapchunks swchunk has allowed values between 2048 and65536 blocks .
Maximum Swap on HP UX 11i V3 The total bytes of swap space manageablesystem on HP UX 11i v3 is swchunk x 1KB x 2147483648 Dump Terms.
Dump unit A thread of execution during dump A dump unit requires itsown set of CPUs dump devices and other resources which are non overlapping with other dump units Reentrancy Capability of a dump driver to issue multiple I Ossimultaneously one I O per HBA port during dump .
Concurrency Capability of a dump driver to issue multiple I Ossimultaneously per HBA port during dump In HP UX 11i v3 this capabilitymeans that the driver can issue I Os simultaneously to multiple devicesunder a given HBA port one I O per device Parallel Dump HP UX 11i v3 dump infrastructure which enables the.
parallelism features Reentrant HBA port or device An HBA port or device controlled by areentrant driver Concurrent HBA port or device An HBA port or device controlled by aconcurrent driver .
Dump Unit Part 1 A Dump Unit is an independent sequential unit ofexecution within the dump process Each dump unit is assigned an exclusive subset of thesystem resources needed to perform the dump including.
CPUs a portion of the physical memory to be dumped and a subset of the configured dump devices The dumpinfrastructure in HP UX 11i v3 automatically partitionssystem resources at dump time into dump units Each dump unit operates sequentially .
Parallelism is achieved by multiple dump units executingin parallel Dump Unit Part 2 A dump device cannot be shared across multiple dump units Multiple reentrant devices can be accessed in parallel only if.
the devices are configured through separate HBA ports Thusall reentrant devices on the same HBA port will be assignedto a single dump unit Each concurrent device can be accessed in parallel Each cantherefore be assigned to a separate dump unit even if.
configured through a single HBA port Multiple dump volumes on a single physical volume will notallow for parallelism Parallelism at dump time can only beachieved across multiple physical devices LUNs Logical volumes configured as dump devices all logical.
volumes which reside on the same physical device LUN areassigned to the same dump unit Dump Options Overview Selective Based on classes uses of memory.
Compressed 5 CPUs per dump unit Mixed compressed non compressed images Parallel concurrent Faster dump with multiple monarchs .
Influenced by memory availability and dump devices HP Integrity Servers only Live dump Crashdump a live system without forced shutdown or panic System stays up running stable.
Offline analysis of system Memory image file Extra load during this save HP Integrity Servers only Dump Parallelism.
I O support during dump is provided via dump drivers and eachconfigureddump driver reports its parallelism capabilities to the dumpinfrastructure Legacy new parallelism feature is not supported.
Reentrant supports parallelism per HBA portConcurrent supports parallelism per dump deviceThese requirements can be distilled into the following formulas forcalculatingthe number of dump units that can be achieved .
CPU Parallelism number of CPUs available at dump time 1 or5 depending on whether or not compression is enabled Device Parallelism number of reentrant dump HBA ports number of concurrent dump devices 1 if there are any legacy Dump Driver Parallelism Capability.
Examples of HP provided dump drivers on HP UX 11 31 fcd Concurrenttd mpt c8xx ciss sasd fclp Reentrant crashconf lDEVICE LOGICAL VOL NAME LUNPATH HANDLE .
1 0x000002 64 0x000002 dev vg00 lvol2 40 0 2 0 0 0 0 4 0 0 0 0x24700... ioscan fNk grep 40 0 2 0 0 0 0 4 0 0 0 fc 0 40 0 2 0 0 0 0 4 0 0 0 fclp CLAIMED INTERFACE HP AD222 60001 PCIe Fibre Channel 2 port 4Gb FC 2 port 1000B T Combo Adapter.
Dump Driver Capability scsimgr get attr a capability H 40 0 2 0 0 0 0 4 0 0 0SCSI ATTRIBUTES FOR CONTROLLER 40 0 2 0 0 0 0 4 0 0 0name capabilitycurrent Boot Dump .
Uncompressed vs CompressedOne Dump Device Uncompressed vs CompressedDump Three Dump Devices Uncompressed vs Compressed.
Dump Legacy Devices Uncompressed Dump Reentrant Uncompressed vs CompressedDump Complex Example Compressed Dump Configuration.
crashconf vCrash dump configuration has been changed since boot CLASS PAGES INCLUDED IN DUMP DESCRIPTION UNUSED 1514754 no by default unused pages.
USERPG 112614 no by default user process pagesBCACHE 26235 no by default buffer cache pagesKCODE 10389 yes forced kernel code pagesUSTACK 1136 yes by default user process stacksFSDATA 40 no forced file system metadata.
KDDATA 386358 yes by default kernel dynamic dataKSDATA 6933 yes by default kernel static dataSUPERPG 21546 no by default unused kernel super pagesTotal pages on system 2080005Total pages included in dump 404816.
Dump compressed ON crashconf c off to turn compression off until reboot crashconf c on to turn compression on until rebootDump Parallel ONDEVICE OFFSET kB SIZE kB LOGICAL VOL NAME .
3 0x000000 2349920 8388608 64 0x000002 dev vg00 lvol23 0x000000 30677856 114688 64 0x000009 dev vg00 v3Dump Dump device configuration mode is config deprecated mode Use crashconf s option to change the mode .
kctune dump compress onTunable Value Expression Changesdump compress on 1 Default Immed crashconf tc off to change tunable to 0 kctune dump compress on 0 crashconf tc on to set tunable to 1.
kctune dump compress on 1 Compressed Dump Algorithm HP UX uses one processor to do all disk writes andfour processors for compression The algorithm for compression is Lempel Ziv .
Welch LZW LZW is a universal lossless data compressionalgorithm simple to implement and has thepotential for very high throughput in hardwareimplementations .
One of the reasons for selecting LZW HP has a license to use it andIt achieves pretty good compression ratio Concurrent Dump Configuration crashconf v.
Crash dump configuration has been changed since boot CLASS PAGES INCLUDED IN DUMP DESCRIPTION UNUSED 1514754 no by default unused pagesUSERPG 112614 no by default user process pages.
BCACHE 26235 no by default buffer cache pagesKCODE 10389 yes forced kernel code pagesUSTACK 1136 yes by default user process stacksFSDATA 40 no forced file system metadataKDDATA 386358 yes by default kernel dynamic data.
KSDATA 6933 yes by default kernel static dataIO Timeout (Seconds) default--- Distribution of logical volume ---PV Name LE on PV PE on PV /dev/dsk/c0t0d0 1904 1904 /dev/dsk/c2t0d0 1904 1904 # /usr/sbin/lvdisplay -v /dev/vg00/swap1--- Logical volumes ---LV Name /dev/vg00/swap1. VG Name /dev/vg00

Related Presentations