Disk Scrub Checksum Error
Contents |
translations. Please help today!Produce and hosting N4F does cost money, please consider a small donation to our project so that we
Wafl Inconsistent User Data Block
can stay offering you the best.We really do need your support!We freenas checksum error encourage all users to add their location (City & Country) and timezone to their profile ! [SOLVED]
Zfs Checksum Errors
Some checksum errors on ramdom disks after scrub Forum rules • NAS4Free Official Site • Set-Up Guide • FAQs • Forum Rules Post Reply Print view 19 posts freenas zpool clear command • Page 1 of 1 erik Starter Posts: 60 Joined: 14 Jul 2014 09:45 Status: Offline [SOLVED] Some checksum errors on ramdom disks after scrub Quote Post #1 by erik » 15 Aug 2014 09:16 I'm running 6 2TB drives in RAID-Z2System is stable and is able to saturate GBit link over sambaCPU is AMD Athlon netapp wafl scan status 64bit, 3 cores low power version12 GByte of memory (non-ECC)power consumption is less then 100 Watt and the PSU is 300 Watt Every weekend I run a scrub.After every scrub there are a small (<10) number of CHEKSUM errors on random disks1: Is this a problem?2: What could be the cause?I recently moved all components to a different case (different routing of cables) and that did not make a change---------------------------- Edit: Summary so you do not have to read the whole thread -----------------------------------------------------------Faulty memory was expected.Running memtest86+ in SMP mode did reveal a faulty memory module.After removal of that memory module all memory configurations with more then 4GByte cause memtest+ to hang.All other computers tested had the same problem so a bug in memtest+ in SMP mode was assumed.The solution was to switch to Round Robin Testing.After the memory test was OK a binary comparison of all files on the zfs pool and a backup made before the data was moved to the ZFS system was
21, 2011 Messages: 95 Thanks Received: 2 Trophy Points: 6 Location: DC I did
Zpool Clear Checksum
a scrub on my volume yesterday, please see the results below. freenas zpool scrub Particularly I am wondering about the checksum errors (81) on the one disk as shown. Does that
Zpool Clear Freenas
mean I need to replace it? I did a zpool clear and it went back to zero... Further, FreeNAS GUI says the volume is "HEALTHY" after the zpool http://www.nas4free.org/forums/viewtopic.php?t=7128 clear. I should have checked it before that, but I didn't...whoops... Code: [root@freenas] ~# zpool status pool: DEC2010 state: ONLINE status: One or more devices has experienced an unrecoverable error. An attempt was made to correct the error. Applications are unaffected. action: Determine if the device needs to be replaced, and clear the errors using 'zpool https://forums.freenas.org/index.php?threads/checksum-errors-after-scrub-on-zpool.12221/ clear' or replace the device with 'zpool replace'. see: http://www.sun.com/msg/ZFS-8000-9P scan: scrub repaired 55K in 16h1m with 0 errors on Thu Apr 4 22:45:13 2013 config: NAME STATE READ WRITE CKSUM DEC2010 ONLINE 0 0 0 raidz1-0 ONLINE 0 0 0 gptid/218a2962-c5d4-11e0-8fd1-0025900d25b0 ONLINE 0 0 0 gptid/dc4978b8-ff57-11e1-88de-0025900d25b0 ONLINE 0 0 81 gptid/22b4bd1e-c5d4-11e0-8fd1-0025900d25b0 ONLINE 0 0 0 gptid/afb125e9-84f7-11e1-b8dd-0025900d25b0 ONLINE 0 0 0 gptid/b3dfe402-006c-11e2-acc4-0025900d25b0 ONLINE 0 0 0 gptid/23cc848c-c5d4-11e0-8fd1-0025900d25b0 ONLINE 0 0 0 raidz1-1 ONLINE 0 0 0 ada6p2.nop ONLINE 0 0 0 ada7p2.nop ONLINE 0 0 0 ada8p2.nop ONLINE 0 0 0 ada9p2.nop ONLINE 0 0 0 errors: No known data errors [root@freenas] ~# zpool clear DEC2010 [root@freenas] ~# zpool status pool: DEC2010 state: ONLINE scan: scrub repaired 55K in 16h1m with 0 errors on Thu Apr 4 22:45:13 2013 config: NAME STATE READ WRITE CKSUM DEC2010 ONLINE 0 0 0 raidz1-0 ONLINE 0 0 0 gptid/218a2962-c5d4-11e0-8fd1-0025900d25b0 ONLINE 0 0 0 gptid/dc4978b8-ff57-11e1-88de-0025900d25b0 ONLINE 0 0 0 gptid/22b4bd1e-c5d4-11e0-8fd1-0025900d25b0 ONLINE 0 0 0 gptid/afb125e9-84f7-11e1-b8dd-0025900d25b0 ONLINE 0 0 0 gptid/b3dfe402-006c-11e2-acc4-0025900d25
a pool scrub was starting: starting scrub of pool 'Disk1' About 3 hours after this, I received http://techblog.danielpellarini.com/sysadmin/how-to-deal-with-checksum-errors-in-zfs-freenas/ another email with an excerpt from the kernel logs: freenas.local kernel log https://forums.freebsd.org/threads/35720/ messages: (ada1:ahcich1:0:0:0): READ_FPDMA_QUEUED. ACB: 60 d8 80 df 4d 40 46 00 00 00 00 00 (ada1:ahcich1:0:0:0): CAM status: ATA Status Error (ada1:ahcich1:0:0:0): ATA status: 51 (DRDY SERV ERR), error: 40 (UNC ) (ada1:ahcich1:0:0:0): RES: 51 40 98 df 4d 40 46 00 00 c0 00 (ada1:ahcich1:0:0:0): Retrying checksum error command (ada1:ahcich1:0:0:0): READ_FPDMA_QUEUED. ACB: 60 d8 80 df 4d 40 46 00 00 00 00 00 (ada1:ahcich1:0:0:0): CAM status: ATA Status Error (ada1:ahcich1:0:0:0): ATA status: 51 (DRDY SERV ERR), error: 40 (UNC ) (ada1:ahcich1:0:0:0): RES: 51 40 98 df 4d 40 46 00 00 c0 00 (ada1:ahcich1:0:0:0): Retrying command (ada1:ahcich1:0:0:0): READ_FPDMA_QUEUED. ACB: 60 d8 80 df 4d 40 46 00 00 disk scrub checksum 00 00 00 (ada1:ahcich1:0:0:0): CAM status: ATA Status Error (ada1:ahcich1:0:0:0): ATA status: 51 (DRDY SERV ERR), error: 40 (UNC ) (ada1:ahcich1:0:0:0): RES: 51 40 98 df 4d 40 46 00 00 c0 00 (ada1:ahcich1:0:0:0): Retrying command (ada1:ahcich1:0:0:0): READ_FPDMA_QUEUED. ACB: 60 d8 80 df 4d 40 46 00 00 00 00 00 (ada1:ahcich1:0:0:0): CAM status: ATA Status Error (ada1:ahcich1:0:0:0): ATA status: 51 (DRDY SERV ERR), error: 40 (UNC ) (ada1:ahcich1:0:0:0): RES: 51 40 98 df 4d 40 46 00 00 c0 00 (ada1:ahcich1:0:0:0): Retrying command (ada1:ahcich1:0:0:0): READ_FPDMA_QUEUED. ACB: 60 d8 80 df 4d 40 46 00 00 00 00 00 (ada1:ahcich1:0:0:0): CAM status: ATA Status Error (ada1:ahcich1:0:0:0): ATA status: 51 (DRDY SERV ERR), error: 40 (UNC ) (ada1:ahcich1:0:0:0): RES: 51 40 98 df 4d 40 46 00 00 c0 00 (ada1:ahcich1:0:0:0): Error 5, Retries exhausted -- End of security output -- Immediately followed by a third email, with a status summary of my pool: Checking status of zfs pools: NAME SIZE ALLOC FREE CAP DEDUP HEALTH ALTROOT Disk1 928G 494G 434G 53% 1.00x ONLINE /mnt pool: Disk1 state: ONLINE status: One
Primer All Books and Articles Community Mailing Lists Forums User Groups Events Q&A (external) Developers Project Ideas SVN Repository Perforce Repository Support Vendors Security Information Bug Reports Submit Bug-report Foundation Monetary Donations Hardware Donations Log in or Sign up The FreeBSD Forums Forums > Base System > Storage > This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More. ZFS scrubbing Discussion in 'Storage' started by Jimmy, Nov 13, 2012. Jimmy Member Messages: 59 Thanks Received: 2 When the handbook talks about scrubbing it mentions additional disk space required for the checksums, isn't scrubbing just checking the stripes and recalculating the parity to see if the parity is accurate? If not why is the checksum needed, isn't it a little redundant when we have parity? Can someone explain in a little more detail how it works? Does each block's inode reference a checksum or is the scrub a completely independent mechanism? Thank you. Jimmy, Nov 13, 2012 #1 Jimmy Member Messages: 59 Thanks Received: 2 Also wanted to ask is scrubbing automatic? And how do we check if a scrub is being performed? Jimmy, Nov 13, 2012 #2 usdmatt Aspiring Daemon Messages: 999 Thanks Received: 318 I'll provide information on what I can, I'm no expert on how raidz actually works. All data written to the zpool is split into records of a certain size (usually 128kB or less). This record is then checksummed and written to disk. The checksum is checked every time that record is read. If you compress data, the compression is done first, then checksummed. If dedupe is enabled, the checksum will be looked up in the DDT (DeDupe Table) to see if that record already exists. I'm not sure how exactly records are written with raidz but obviously data is not always stripped with parity - many people use mirrored pools or even single disk pools. In a single disk pool data can still be recovered from checksum error if you tell ZFS to store more than one copy of each record. From what I can find on the net I think that data read from raidz is validated with the checksum. The parity is only read if data needs to be reconstructed although it's not very clear. Scrub is not automatic. You can start a scrub, and then check progress with the commands below. If a scrub is in progress the status command will tell you and give the progress, speed and estimated time to completion. I believe there are also periodic scripts that can be enabled to do this automatically at a set interval. Code: # zpool scrub pool # zpool status