Preclear plugin

JorgeB · March 12, 2016

Please run the command below and send the the file preclear_beta.txt created on your flash drive:

Attached.

gfjardim · March 12, 2016

Please run the command below and send the the file preclear_beta.txt created on your flash drive:

Attached.

Err, wrong command...

Could you please try again?

bash -x /usr/local/emhttp/plugins/preclear.disk.beta/script/preclear_disk.sh /dev/sde >/boot/preclear_beta.txt 2>&1

Thanks

JorgeB · March 12, 2016

Err, wrong command...

Could you please try again?

No problem.

preclear_beta.txt

gfjardim · March 12, 2016

Err, wrong command...

Could you please try again?

No problem.

Thank you very much. Please update the plugin and test it again.

saarg · March 12, 2016

So far no problems starting the preclear on two 4TB disks.

Was it stupid to choose 5 cycles? Just joking

JorgeB · March 12, 2016

Thank you very much. Please update the plugin and test it again.

Updated to 2016.03.12 and still get same error, let me know of need any more info.

gfjardim · March 12, 2016

Thank you very much. Please update the plugin and test it again.

Updated to 2016.03.12 and still get same error, let me know of need any more info.

Please do the test again and send me the file.

JorgeB · March 12, 2016

Please do the test again and send me the file.

Here you go.

preclear_beta.txt

mr-hexen · March 12, 2016

@danioj,

The error noted in your tests (well done, btw) with regards to running the old preclear script are known issues w/ 6.2.

gfjardim · March 12, 2016

Please do the test again and send me the file.

Here you go.

Apparently it didn't update correctly. Try to update to 2016.03.12a and then test it.

JorgeB · March 12, 2016

Apparently it didn't update correctly. Try to update to 2016.03.12a and then test it.

Working now

Will report if all went well.

P.S. - looks like there's some wrong info on the status display:

Time elapsed: 0:02:42 | Write speed: 159.62s, | Average speed: 83MB/s

Plugin page:

Zeroing: 78% @ 296.065s, MB/s (0:04:58)

mr-hexen · March 12, 2016

@gfjardim,

All my (eventual) testing will be done via 6.1.9 as I'm seeing too many issues reported by folks upgrading to 6.2b18 and I have only 1 server

At least this will add a data point for backwards compatibility.

mr-hexen · March 12, 2016

Apparently it didn't update correctly. Try to update to 2016.03.12a and then test it.

Working now

Will report if all went well.

P.S. - looks like there's some wrong info on the status display:
Time elapsed: 0:02:42 | Write speed: 159.62s, | Average speed: 83MB/s  
Plugin page:
Zeroing: 78% @ 296.065s, MB/s (0:04:58)

I noticed this too on reported tests (current speed) is in seconds, not mb/s.

JorgeB · March 12, 2016

3) Once it's over, please run a Verify All the Disk operation;

verify successful

4) If successful, please run both Test and Run the post-read verify operations from the older plugin;

can't at the moment because old plugin does not work on 6.2

5) If possible, assign the precleared disk to the array and start a Check Parity operation.

successfully assigned and parity check finished with 0 sync errors

gfjardim · March 12, 2016

3) Once it's over, please run a Verify All the Disk operation;

verify successful

4) If successful, please run both Test and Run the post-read verify operations from the older plugin;

can't at the moment because old plugin does not work on 6.2

5) If possible, assign the precleared disk to the array and start a Check Parity operation.

successfully assigned and parity check finished with 0 sync errors

This is the greatest test. If unRAID accept it as precleared and a parity check run without sync errors, then the preclear is successful.

Thank a lot for your input, johnnie.black!

RobJ · March 12, 2016

Some thoughts -

* I've long wrestled with this, trying to find a way forward, to clear the air, to bring Preclear functionality up to date in plugin form, and I have to say your efforts here are wonderful, near miraculous, so unexpected, but greatly appreciated. I am absolutely positive that Joe never intended to hold Preclear functionality hostage, but that is what effectively had happened. Like most developers, Joe really wanted complete control over his code, and since he's a Linux veteran and command-line oriented, never appreciated the overwhelming desire for GUI style tools, visual plugins over scripts. Plus he got very busy with a new job, didn't have time to keep up with all the changes. Except for one small change for v6, he never even had time for changes with code suggested by various other users.

* I had recently derived patch files (from diff -u, for patch), taking the original script of 1.15, first back to 1.14, then to the fast Preclear, then to the latest with changes by gfjardim, then include a small change of mine plus a prominent statement included that would say something like "Because script modifications by various unRAID community members are included, this Preclear script is not authorized or supported by the original author, Joe L." I had hoped that that would make it at least ethically acceptable, and provide a way forward, using a line of mods/patches for all the modifications to come. Thankfully, that is no longer needed, completely superseded by your rewrite. A cloud has been lifted from the community.

* Joe didn't want changes that weren't his, that he couldn't support, but as a support issue, that never really held water. He was so good with sed and diff and bash scripting in general, he would quickly grasp any small changes from us. And besides, it's the whole community supporting Preclear. Joe hasn't had to support it in a long time.

* There had been suggestions to rewrite Preclear functionality, but that was NEVER plausible, a complete non-starter. It would be completely illogical to rewrite something that works so well, and is well-tested, well-understood, especially with so many other things to work on. It was going to be a LONG time before it made sense for LimeTech to take time to write something similar, something that duplicated the current functionality. (That's why your efforts are so unexpected, delightfully illogical, near miraculous! We owe you a large debt of gratitude.)

* The small section that reports drive issues at the close of the Preclear session has long had an error that causes confusion, see this example -

** Changed attributes in files: /tmp/smart_start_sdi  /tmp/smart_finish_sdi
                ATTRIBUTE   NEW_VAL OLD_VAL FAILURE_THRESHOLD STATUS      RAW_VALUE
      Raw_Read_Error_Rate =   115     114            6        ok          94152625
          Seek_Error_Rate =    52      52           30        near_thresh 1322898620909
         Spin_Retry_Count =   100     100           97        near_thresh 0
         End-to-End_Error =   100     100           99        near_thresh 0
          High_Fly_Writes =    96      97            0        ok          4
  Airflow_Temperature_Cel =    67      66           45        near_thresh 33
      Temperature_Celsius =    33      34            0        ok          33
   Hardware_ECC_Recovered =    49      43            0        ok          94152625
No SMART attributes are FAILING_NOW

None of the 4 "near_thresh" lines should have been reported, and probably not any of the others either (others opinions may differ). They aren't a real concern, so listing them causes confusion to less experienced users. I'm not saying don't list real problems, but I would prefer to *only* list those of concern. This is a good example where we can now propose needed changes, and hope to see them happen!

* I had long wanted to propose a 'strategy' feature, where the user could specify the testing and clearing strategy used. I had thought to add an option with perhaps "-p string" (-p for plan, couldn't use -s as it was already in use, before your rewrite). 'String' would consist of an ordered list of digits or letters representing operations, such as 0=preread, 1=zeroing, 2=old postread, 3=fast postread, 4=smart long test, 5=badblocks -n, and 6=badblocks -w. SMART reports would be obtained and checked before starting and after every operation. A Preclear (with fast) like the current default, would default to "-p 013". But you could now easily create your own favorite Preclearing process, such as "-p 613" or "-p 413" or "-p 046135313". Or you could do a quick zeroing (-p 1) or quick Preclear (-p 13), or if you don't need zeroing just testing (-p 46 or -p 45). Seems simple and intuitive for users. It's actually more complicated to implement though, as there are other operations, such as write signature, test for signature, reads with random head movement, test for all zeros (only after a zeroing or operation that preserved a previous zeroing, like reads and smart tests), etc. The current postreads are actually 3 operations in one, I believe. Just ideas for the future ...

gfjardim · March 12, 2016

Some thoughts -

* I've long wrestled with this, trying to find a way forward, to clear the air, to bring Preclear functionality up to date in plugin form, and I have to say your efforts here are wonderful, near miraculous, so unexpected, but greatly appreciated. I am absolutely positive that Joe never intended to hold Preclear functionality hostage, but that is what effectively had happened. Like most developers, Joe really wanted complete control over his code, and since he's a Linux veteran and command-line oriented, never appreciated the overwhelming desire for GUI style tools, visual plugins over scripts. Plus he got very busy with a new job, didn't have time to keep up with all the changes. Except for one small change for v6, he never even had time for changes with code suggested by various other users.

* I had recently derived patch files (from diff -u, for patch), taking the original script of 1.15, first back to 1.14, then to the fast Preclear, then to the latest with changes by gfjardim, then include a small change of mine plus a prominent statement included that would say something like "Because script modifications by various unRAID community members are included, this Preclear script is not authorized or supported by the original author, Joe L." I had hoped that that would make it at least ethically acceptable, and provide a way forward, using a line of mods/patches for all the modifications to come. Thankfully, that is no longer needed, completely superseded by your rewrite. A cloud has been lifted from the community.

I do understand Joe L. position, and I don't blame him. Last year's end I got real busy and when I got back three months later a parallel version of Unassigned Devices was being developed by dlandon. I got real mad about him at first, but then I realize he was doing a better job supporting it than I do, so I GPL'ed the code and let him take care of it. In fact, now I thank him to do that work for me ::) .

* Joe didn't want changes that weren't his, that he couldn't support, but as a support issue, that never really held water. He was so good with sed and diff and bash scripting in general, he would quickly grasp any small changes from us. And besides, it's the whole community supporting Preclear. Joe hasn't had to support it in a long time.

I do agree. In a community like ours, a lot of users do want to help, sometimes with feedback, sometimes helping other less experienced users. You don't need to do all by yourself. Delegation and cooperation is the key to maintain an open source project.

* There had been suggestions to rewrite Preclear functionality, but that was NEVER plausible, a complete non-starter. It would be completely illogical to rewrite something that works so well, and is well-tested, well-understood, especially with so many other things to work on. It was going to be a LONG time before it made sense for LimeTech to take time to write something similar, something that duplicated the current functionality. (That's why your efforts are so unexpected, delightfully illogical, near miraculous! We owe you a large debt of gratitude.)

Thank you a lot for your kind words! It began with ljm42 asking me to include the script within the plugin, because many users were having problems with two versions of it, the disks.cfg issue etc... Then I started to patch the original code, but when I fixed one thing, another one got borked, so after some time, I do realize that a major rewrite was needed.

* The small section that reports drive issues at the close of the Preclear session has long had an error that causes confusion, see this example -
** Changed attributes in files: /tmp/smart_start_sdi  /tmp/smart_finish_sdi
                ATTRIBUTE   NEW_VAL OLD_VAL FAILURE_THRESHOLD STATUS      RAW_VALUE
      Raw_Read_Error_Rate =   115     114            6        ok          94152625
          Seek_Error_Rate =    52      52           30        near_thresh 1322898620909
         Spin_Retry_Count =   100     100           97        near_thresh 0
         End-to-End_Error =   100     100           99        near_thresh 0
          High_Fly_Writes =    96      97            0        ok          4
  Airflow_Temperature_Cel =    67      66           45        near_thresh 33
      Temperature_Celsius =    33      34            0        ok          33
   Hardware_ECC_Recovered =    49      43            0        ok          94152625
No SMART attributes are FAILING_NOW
None of the 4 "near_thresh" lines should have been reported, and probably not any of the others either (others opinions may differ). They aren't a real concern, so listing them causes confusion to less experienced users. I'm not saying don't list real problems, but I would prefer to *only* list those of concern. This is a good example where we can now propose needed changes, and hope to see them happen!

I haven't started that part yet, so I'll need to study this a little more. I definitely will make things more readable.

* I had long wanted to propose a 'strategy' feature, where the user could specify the testing and clearing strategy used. I had thought to add an option with perhaps "-p string" (-p for plan, couldn't use -s as it was already in use, before your rewrite). 'String' would consist of an ordered list of digits or letters representing operations, such as 0=preread, 1=zeroing, 2=old postread, 3=fast postread, 4=smart long test, 5=badblocks -n, and 6=badblocks -w. SMART reports would be obtained and checked before starting and after every operation. A Preclear (with fast) like the current default, would default to "-p 013". But you could now easily create your own favorite Preclearing process, such as "-p 613" or "-p 413" or "-p 046135313". Or you could do a quick zeroing (-p 1) or quick Preclear (-p 13), or if you don't need zeroing just testing (-p 46 or -p 45). Seems simple and intuitive for users. It's actually more complicated to implement though, as there are other operations, such as write signature, test for signature, reads with random head movement, test for all zeros (only after a zeroing or operation that preserved a previous zeroing, like reads and smart tests), etc. The current postreads are actually 3 operations in one, I believe. Just ideas for the future ...

If could be done, but I think it would make things messier. The good thing is that if we want to use badblocks instead of the current method, we could just write a new function and then add a option. It's a thing we should evaluate, tho. The next crucial step is to add the SMART monitoring function, and then we could evaluate the use of badblocks etc...

In my code, post-reads are just one operation: read a bunch of sectors, sum them and then verify if the result is equal to 0. All reads are defaulted to hammer the read/write heads with Joe L. method of reading three random sectors, than the first and the last one, but it can be disabled from the command line.

Again, thanks a lot for this message, I really appreciate your ideas and your feedback.

ljm42 · March 12, 2016

This is amazing gfjardim, thank you

saarg · March 13, 2016

The new beta preclear failed for me when verifying the unRAID's Preclear signature. See below.

This was on two Seagate 4TB that have not been used before. I formatted them with XFS in unassigned devices and transferred some files to them before starting the preclear.

This was tested on 6.1.9.

###########################################################################################
#                                                                                         #
#                        unRAID Server Pre-Clear of disk /dev/sdd                         #
#                       Cycle 1 of 1, partition start on sector 64.                       #
#                                                                                         #
#                                                                                         #
#                                                                                         #
#   Step 1 of 5 - Pre-read verification:                                         SUCCESS  #
#   Step 2 of 5 - Zeroing the disk:                                              SUCCESS  #
#   Step 3 of 5 - Writing unRAID's Preclear signature:                           SUCCESS  #
#   Step 4 of 5 - Verifying unRAID's Preclear signature:                            FAIL  #
#                                                                                         #
#                                                                                         #
#                                                                                         #
#                                                                                         #
#                                                                                         #
#                                                                                         #
#                                                                                         #
###########################################################################################
#                 Cycle elapsed time: 16:36:40 | Total elapsed time: 16:36:40             #
###########################################################################################
--> FAIL: unRAID's Preclear signature not valid.
root@unXeon:/usr/local/emhttp#

Syslog. I don't think the MCE errors have anything to do with the signature errors as it happened during the pre-read.

Mar 12 12:37:32 unXeon udevd[23747]: inotify_add_watch(6, /dev/sdc1, 10) failed: No such file or directory
Mar 12 12:37:32 unXeon kernel: sdc: sdc1
Mar 12 12:37:58 unXeon udevd[24100]: inotify_add_watch(6, /dev/sdd1, 10) failed: No such file or directory
Mar 12 12:37:58 unXeon kernel: sdd: sdd1
Mar 12 12:38:11 unXeon kernel: XFS (sdc1): Mounting V4 Filesystem
Mar 12 12:38:11 unXeon kernel: XFS (sdc1): Ending clean mount
Mar 12 12:38:19 unXeon kernel: XFS (sdd1): Mounting V4 Filesystem
Mar 12 12:38:19 unXeon kernel: XFS (sdd1): Ending clean mount
Mar 12 12:38:34 unXeon sshd[24818]: Accepted none for root from 192.168.1.113 port 50037 ssh2
Mar 12 12:40:55 unXeon kernel: XFS (sdc1): Unmounting Filesystem
Mar 12 12:41:12 unXeon kernel: XFS (sdd1): Unmounting Filesystem
Mar 12 12:46:04 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 13:04:59 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 13:14:42 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 13:25:18 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 13:27:17 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 13:31:41 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 13:32:49 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 13:37:28 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 13:41:07 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 13:46:05 unXeon kernel: mce_notify_irq: 1 callbacks suppressed
Mar 12 13:46:05 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 13:47:02 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 13:48:22 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 13:49:56 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 14:01:34 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 14:05:12 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 14:09:03 unXeon kernel: mce_notify_irq: 1 callbacks suppressed
Mar 12 14:09:03 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 14:17:22 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 14:30:34 unXeon kernel: mce_notify_irq: 1 callbacks suppressed
Mar 12 14:30:34 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 14:31:30 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 14:34:52 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 14:35:23 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 14:38:36 unXeon kernel: mce_notify_irq: 1 callbacks suppressed
Mar 12 14:38:36 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 14:41:16 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 14:42:35 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 15:00:54 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 15:15:39 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 15:30:51 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 12 15:52:29 unXeon kernel: mce: [Hardware Error]: Machine check events logged
Mar 13 00:33:24 unXeon dhcpcd[2229]: br0: renewing lease of 192.168.1.7
Mar 13 00:33:24 unXeon dhcpcd[2229]: br0: rebind in 32400 seconds, expire in 43200 seconds
Mar 13 00:33:24 unXeon dhcpcd[2229]: br0: sending REQUEST (xid 0xb0ad89e1), next in 3.6 seconds
Mar 13 00:33:24 unXeon dhcpcd[2229]: br0: acknowledged 192.168.1.7 from 192.168.1.1
Mar 13 00:33:24 unXeon dhcpcd[2229]: br0: leased 192.168.1.7 for 86400 seconds
Mar 13 00:33:24 unXeon dhcpcd[2229]: br0: renew in 39262 seconds, rebind in 71662 seconds
Mar 13 00:33:24 unXeon dhcpcd[2229]: br0: writing lease `/var/lib/dhcpcd/dhcpcd-br0.lease'
Mar 13 00:33:24 unXeon dhcpcd[2229]: br0: IP address 192.168.1.7/24 already exists
Mar 13 00:33:24 unXeon dhcpcd[2229]: br0: executing `/lib/dhcpcd/dhcpcd-run-hooks' RENEW
Mar 13 00:33:24 unXeon dhcpcd[2229]: br0: ARP announcing 192.168.1.7 (1 of 2), next in 2.0 seconds
Mar 13 00:33:26 unXeon dhcpcd[2229]: br0: ARP announcing 192.168.1.7 (2 of 2)
Mar 13 05:18:00 unXeon udevd[8232]: timeout 'ata_id --export /dev/sdd'
Mar 13 05:18:01 unXeon udevd[8232]: timeout: killing 'ata_id --export /dev/sdd' [8233]
Mar 13 05:18:02 unXeon udevd[8232]: timeout: killing 'ata_id --export /dev/sdd' [8233]
Mar 13 05:18:03 unXeon udevd[8232]: timeout: killing 'ata_id --export /dev/sdd' [8233]
Mar 13 05:18:04 unXeon udevd[8232]: timeout: killing 'ata_id --export /dev/sdd' [8233]
Mar 13 05:18:05 unXeon udevd[8232]: timeout: killing 'ata_id --export /dev/sdd' [8233]
Mar 13 05:18:06 unXeon udevd[8232]: timeout: killing 'ata_id --export /dev/sdd' [8233]
Mar 13 05:18:07 unXeon udevd[8232]: timeout: killing 'ata_id --export /dev/sdd' [8233]
Mar 13 05:18:08 unXeon udevd[8232]: timeout: killing 'ata_id --export /dev/sdd' [8233]
Mar 13 05:18:09 unXeon udevd[8232]: timeout: killing 'ata_id --export /dev/sdd' [8233]
Mar 13 05:18:10 unXeon udevd[8232]: timeout: killing 'ata_id --export /dev/sdd' [8233]
Mar 13 05:18:10 unXeon udevd[8232]: 'ata_id --export /dev/sdd' [8233] terminated by signal 9 (Killed)
Mar 13 05:18:10 unXeon udevd[8232]: timeout 'scsi_id --export --whitelisted -d /dev/sdd'
Mar 13 05:38:43 unXeon udevd[23418]: timeout 'ata_id --export /dev/sdc'
Mar 13 05:38:44 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:38:45 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:38:46 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:38:47 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:38:48 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:38:49 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:38:50 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:38:51 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:38:52 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:38:53 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:38:54 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:38:55 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:38:56 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:38:57 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:38:58 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:38:59 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:00 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:01 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:02 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:03 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:04 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:05 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:06 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:07 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:08 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:09 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:10 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:11 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:12 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:13 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:14 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:15 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:16 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:17 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:18 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:19 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:20 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:21 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:22 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:23 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:24 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:25 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:26 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:27 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:28 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:29 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:30 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:31 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:32 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:33 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:34 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:35 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:36 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:37 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:38 unXeon udevd[23418]: timeout: killing 'ata_id --export /dev/sdc' [23419]
Mar 13 05:39:38 unXeon udevd[23418]: 'ata_id --export /dev/sdc' [23419] terminated by signal 9 (Killed)
Mar 13 05:39:38 unXeon udevd[23418]: timeout 'scsi_id --export --whitelisted -d /dev/sdc'

JorgeB · March 13, 2016

Since last post precleared 4 or 5 more disks, all were added to the array with no issues, parity sync always completed with 0 errors, although for now only tested on small SSDs.

Having an issue with not finding all available disks, /dev/sdo appears as unassigned on main page but it’s not available for preclear, preclear_beta.txt attached.

preclear_beta.txt

RobJ · March 13, 2016

The new beta preclear failed for me when verifying the unRAID's Preclear signature. See below.

This was on two Seagate 4TB that have not been used before. I formatted them with XFS in unassigned devices and transferred some files to them before starting the preclear.

This was tested on 6.1.9.
#   Step 1 of 5 - Pre-read verification:                                         SUCCESS  #
#   Step 2 of 5 - Zeroing the disk:                                              SUCCESS  #
#   Step 3 of 5 - Writing unRAID's Preclear signature:                           SUCCESS  #
#   Step 4 of 5 - Verifying unRAID's Preclear signature:                            FAIL  #
Syslog. I don't think the MCE errors have anything to do with the signature errors as it happened during the pre-read.

The problem is, with those MCE's, it's hard to trust the machine, or anything it reports. Those MCE's really need to be resolved first, then see if Preclear works. I'd start with a long Memtest, and check the memory timings.

mr-hexen · March 13, 2016

The new beta preclear failed for me when verifying the unRAID's Preclear signature. See below.

This was on two Seagate 4TB that have not been used before. I formatted them with XFS in unassigned devices and transferred some files to them before starting the preclear.

This was tested on 6.1.9.
#   Step 1 of 5 - Pre-read verification:                                         SUCCESS  #
#   Step 2 of 5 - Zeroing the disk:                                              SUCCESS  #
#   Step 3 of 5 - Writing unRAID's Preclear signature:                           SUCCESS  #
#   Step 4 of 5 - Verifying unRAID's Preclear signature:                            FAIL  #
Syslog. I don't think the MCE errors have anything to do with the signature errors as it happened during the pre-read.
The problem is, with those MCE's, it's hard to trust the machine, or anything it reports. Those MCE's really need to be resolved first, then see if Preclear works. I'd start with a long Memtest, and check the memory timings.

MCE errors are a PAIN in the ASS to track down. I used to get them from time to time on v5 and noticed no issues related to them.

Everything I read on the matter pointed to requiring a BIOS update, to which did nothing. Memtest came back clean.

In the end, they went away when I upgraded to v6. So software CAN cause these to happen, likely an incompatibility with the kernel and my hardware (which is not that unique).

saarg · March 13, 2016

The new beta preclear failed for me when verifying the unRAID's Preclear signature. See below.

This was on two Seagate 4TB that have not been used before. I formatted them with XFS in unassigned devices and transferred some files to them before starting the preclear.

This was tested on 6.1.9.
#   Step 1 of 5 - Pre-read verification:                                         SUCCESS  #
#   Step 2 of 5 - Zeroing the disk:                                              SUCCESS  #
#   Step 3 of 5 - Writing unRAID's Preclear signature:                           SUCCESS  #
#   Step 4 of 5 - Verifying unRAID's Preclear signature:                            FAIL  #
Syslog. I don't think the MCE errors have anything to do with the signature errors as it happened during the pre-read.
The problem is, with those MCE's, it's hard to trust the machine, or anything it reports. Those MCE's really need to be resolved first, then see if Preclear works. I'd start with a long Memtest, and check the memory timings.

If the problems were from the MCE errors, they should have been around where preclear writes the signature. I have checked the mcelog and it is related to one of my VM's, so I'm pretty sure it has nothing to do with preclear's error.

I have seen this errors before also, but there have never ever been a crash or any other problems with this computer. So I trust it

Anyway, I moved one of the disks to my soon to be main server and started a preclear there also just to rule out the mce's

JorgeB · March 13, 2016

Just finished preclearing a 3TB WD Green, this disk was previously used on another server so skipped pre and post reads, but failed verifying signature:

###########################################################################################
#                                                                                         #
#                        unRAID Server Pre-Clear of disk /dev/sdn                         #
#                       Cycle 1 of 1, partition start on sector 64.                       #
#                                                                                         #
#                                                                                         #
#                                                                                         #
#   Step 1 of 3 - Zeroing the disk:                                              SUCCESS  #
#   Step 2 of 3 - Writing unRAID's Preclear signature:                           SUCCESS  #
#   Step 3 of 3 - Verifying unRAID's Preclear signature:                            FAIL  #
#                                                                                         #
#                                                                                         #
#                                                                                         #
#                                                                                         #
#                                                                                         #
#                                                                                         #
#                                                                                         #
#                                                                                         #
###########################################################################################
#                 Cycle elapsed time: 7:10:21 | Total elapsed time: 7:10:21               #
###########################################################################################
--> FAIL: unRAID's Preclear signature not valid.
root@Tower2:/usr/local/emhttp#

This server is still on V6.1 so I can also test with the old plugin:

root@Tower2:/boot# preclear_disk.sh -t /dev/sdn
Pre-Clear unRAID Disk /dev/sdn
################################################################## 1.15
Model Family:     Western Digital Caviar Green (AF, SATA 6Gb/s)
Device Model:     WDC WD30EZRX-00D8PB0
Serial Number:    WD-WCC4N5DDZX6V
LU WWN Device Id: 5 0014ee 20bdc27d7
Firmware Version: 80.00A80
User Capacity:    3,000,592,982,016 bytes [3.00 TB]

Disk /dev/sdn: 3000.6 GB, 3000592982016 bytes
255 heads, 63 sectors/track, 364801 cylinders, total 5860533168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disk identifier: 0x00000000

   Device Boot      Start         End      Blocks   Id  System
/dev/sdn1              64   732566645   366283291    0  Empty
########################################################################
failed test 2 00000 00000 00000 00000
failed test 3 00000 00000 00000 00000
failed test 6
========================================================================1.15
==
== Disk /dev/sdn is NOT precleared
== 64 732566582 5860533104
============================================================================

saarg · March 13, 2016

Good to know it's not only me

Preclear plugin

Recommended Posts

Link to comment

Top Posters In This Topic

Popular Days

Top Posters In This Topic

Popular Days

Popular Posts

gfjardim

Squid

CHBMB

Posted Images

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Link to comment

Join the conversation