From: Jon H. <jd_...@ya...> - 2008-11-12 15:44:00
|
What is wrong with this: DEVICESCAN -d sat -s (S/../.././06|L/../.././15) -m my...@em... -W 2,40,45 -M test -o on -f -H -l selftest -l error -C 197 -U 198 the short test works ok, but the Long test never seeems to run.. originally i wanted this to run only on sunday but changed it to daily to test it... ----------------------- N: Jon Hardcastle E: Jon@eHardcastle.com '..Be fearful when others are greedy, and be greedy when others are fearful.' ----------------------- |
From: Bruce A. <ba...@gr...> - 2008-11-12 16:07:08
|
Jon, I think this should work. Could you mail the output of the smartd logging, something like 'grep smartd /var/log/messages' or equivalent for your system? Smartd should report when it sees a regular expression match. Cheers, Bruce On Wed, 12 Nov 2008, Jon Hardcastle wrote: > What is wrong with this: > > DEVICESCAN -d sat -s (S/../.././06|L/../.././15) -m my...@em... -W 2,40,45 -M test -o on -f -H -l selftest -l error -C 197 -U 198 > > the short test works ok, but the Long test never seeems to run.. originally i wanted this to run only on sunday but changed it to daily to test it... > > > ----------------------- > N: Jon Hardcastle > E: Jon@eHardcastle.com > '..Be fearful when others are greedy, and be greedy when others are fearful.' > ----------------------- > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's challenge > Build the coolest Linux based applications with Moblin SDK & win great prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > _______________________________________________ > Smartmontools-support mailing list > Sma...@li... > https://lists.sourceforge.net/lists/listinfo/smartmontools-support > |
From: Christian F. <Chr...@t-...> - 2008-11-12 16:11:30
|
Jon Hardcastle wrote: > What is wrong with this: > > DEVICESCAN -d sat -s (S/../.././06|L/../.././15) -m [...] > the short test works ok, but the Long test never seeems to run.. > originally i wanted this to run only on sunday but changed it to daily > to test it... > Please try 'smartd -q showtests' to check the evaluation of the regexp. Cheers, Christian |
From: Jon H. <jd_...@ya...> - 2008-11-12 16:42:49
|
--- On Wed, 12/11/08, Christian Franke <Chr...@t-...> wrote: > From: Christian Franke <Chr...@t-...> > Subject: Re: [smartmontools-support] Anyone tell me.. > To: Jon@eHardcastle.com, sma...@li... > Date: Wednesday, 12 November, 2008, 4:11 PM > Jon Hardcastle wrote: > > What is wrong with this: > > > > DEVICESCAN -d sat -s (S/../.././06|L/../.././15) -m > [...] > > the short test works ok, but the Long test never > seeems to run.. > > originally i wanted this to run only on sunday but > changed it to daily > > to test it... > > > > Please try 'smartd -q showtests' to check the > evaluation of the regexp. > > Cheers, > Christian This test seems to show that the test is due... SNIPPET Next scheduled self tests (at most 5 of each type per device): Device: /dev/sda, will do test 1 of type S at Thu Nov 13 06:13:13 2008 GMT Device: /dev/sdb, will do test 1 of type S at Thu Nov 13 06:13:13 2008 GMT Device: /dev/sda, will do test 1 of type L at Thu Nov 13 15:13:13 2008 GMT Device: /dev/sdb, will do test 1 of type L at Thu Nov 13 15:13:13 2008 GMT could the offline testing stop the long test from happening? Also, Bruce - messages show very little regarding smartd.. but i will show you upto last restart of smartd ---------- Nov 10 16:11:19 subversion smartd[30248]: smartd received signal 15: Terminated Nov 10 16:11:19 subversion smartd[30248]: smartd is exiting (exit status 0) Nov 10 16:11:19 subversion smartd[30614]: smartd version 5.37 [i386-redhat-linux-gnu] Copyright (C) 2002-6 Bruce Allen Nov 10 16:11:19 subversion smartd[30614]: Home page is http://smartmontools.sourceforge.net/#012 Nov 10 16:11:19 subversion smartd[30614]: Opened configuration file /etc/smartd.conf Nov 10 16:11:19 subversion smartd[30614]: Configuration file /etc/smartd.conf was parsed, found DEVICESCAN, scanning devices Nov 10 16:11:19 subversion smartd[30614]: Device: /dev/sda, opened Nov 10 16:11:19 subversion smartd[30614]: Device: /dev/sda, found in smartd database. Nov 10 16:11:19 subversion smartd[30614]: Device: /dev/sda, enabled SMART Automatic Offline Testing. Nov 10 16:11:19 subversion smartd[30614]: Device: /dev/sda, is SMART capable. Adding to "monitor" list. Nov 10 16:11:19 subversion smartd[30614]: Device: /dev/sdb, opened Nov 10 16:11:19 subversion smartd[30614]: Device: /dev/sdb, not found in smartd database. Nov 10 16:11:20 subversion smartd[30614]: Device: /dev/sdb, enabled SMART Automatic Offline Testing. Nov 10 16:11:20 subversion smartd[30614]: Device: /dev/sdb, is SMART capable. Adding to "monitor" list. Nov 10 16:11:20 subversion smartd[30614]: Monitoring 2 ATA and 0 SCSI devices Nov 10 16:11:20 subversion smartd[30614]: Executing test of mail to jon...@cl... ... Nov 10 16:11:20 subversion smartd[30614]: Test of mail to jon...@cl...: successful Nov 10 16:11:20 subversion smartd[30614]: Device: /dev/sda, initial Temperature is 26 Celsius Nov 10 16:11:21 subversion smartd[30614]: Executing test of mail to jon...@cl... ... Nov 10 16:11:21 subversion smartd[30614]: Test of mail to jon...@cl...: successful Nov 10 16:11:21 subversion smartd[30614]: Device: /dev/sdb, initial Temperature is 25 Celsius Nov 10 16:11:21 subversion smartd[30628]: smartd has fork()ed into background mode. New PID=30628. Nov 11 06:11:22 subversion smartd[30628]: Device: /dev/sda, starting scheduled Short Self-Test. Nov 11 06:11:25 subversion smartd[30628]: Device: /dev/sdb, starting scheduled Short Self-Test. Nov 11 10:11:21 subversion smartd[30628]: Device: /dev/sda, Temperature changed -1 Celsius to 25 Celsius (Min/Max 25!/26) Nov 12 06:11:21 subversion smartd[30628]: Device: /dev/sda, starting scheduled Short Self-Test. Nov 12 06:11:22 subversion smartd[30628]: Device: /dev/sdb, Temperature changed +2 Celsius to 27 Celsius (Min/Max 25/27!) Nov 12 06:11:25 subversion smartd[30628]: Device: /dev/sdb, starting scheduled Short Self-Test. ------------- |
From: Christian F. <Chr...@t-...> - 2008-11-12 21:28:15
|
Jon Hardcastle wrote: >> >> Please try 'smartd -q showtests' to check the >> evaluation of the regexp. >> >> Cheers, >> Christian >> > > This test seems to show that the test is due... > > SNIPPET > > Next scheduled self tests (at most 5 of each type per device): > Device: /dev/sda, will do test 1 of type S at Thu Nov 13 06:13:13 2008 GMT > Device: /dev/sdb, will do test 1 of type S at Thu Nov 13 06:13:13 2008 GMT > Device: /dev/sda, will do test 1 of type L at Thu Nov 13 15:13:13 2008 GMT > Device: /dev/sdb, will do test 1 of type L at Thu Nov 13 15:13:13 2008 GMT > > This looks good. > could the offline testing stop the long test from happening? > No, smartd skips a test only if the "self test execution status" byte indicates another running self-test. This should not happen if the offline data collection is running. > Also, Bruce - messages show very little regarding smartd.. but i will show you upto last restart of smartd > > Did smartd print any message between 15:00 and 16:00 ? Cheers, Christian |
From: Jon H. <jd_...@ya...> - 2008-11-14 09:36:13
|
--- On Wed, 12/11/08, Christian Franke <Chr...@t-...> wrote: > From: Christian Franke <Chr...@t-...> > Subject: Re: [smartmontools-support] Anyone tell me.. > To: Jon@eHardcastle.com > Cc: sma...@li..., "Bruce Allen" <ba...@gr...> > Date: Wednesday, 12 November, 2008, 9:27 PM > Jon Hardcastle wrote: > >> > >> Please try 'smartd -q showtests' to check > the > >> evaluation of the regexp. > >> > >> Cheers, > >> Christian > >> > > > > This test seems to show that the test is due... > > > > SNIPPET > > > > Next scheduled self tests (at most 5 of each type per > device): > > Device: /dev/sda, will do test 1 of type S at Thu Nov > 13 06:13:13 2008 GMT > > Device: /dev/sdb, will do test 1 of type S at Thu Nov > 13 06:13:13 2008 GMT > > Device: /dev/sda, will do test 1 of type L at Thu Nov > 13 15:13:13 2008 GMT > > Device: /dev/sdb, will do test 1 of type L at Thu Nov > 13 15:13:13 2008 GMT > > > > > > This looks good. > > > > could the offline testing stop the long test from > happening? > > > > No, smartd skips a test only if the "self test > execution status" byte > indicates another running self-test. This should not happen > if the > offline data collection is running. > > > > Also, Bruce - messages show very little regarding > smartd.. but i will show you upto last restart of smartd > > > > > > Did smartd print any message between 15:00 and 16:00 ? > > Cheers, > Christian I restarted smartd at 16:00 but for completion.. here is the log entried prior to 16:00 Nov 9 06:33:00 subversion smartd[30248]: Device: /dev/sda, starting scheduled Short Self-Test. Nov 9 06:33:03 subversion smartd[30248]: Device: /dev/sdb, starting scheduled Short Self-Test. Nov 10 06:32:59 subversion smartd[30248]: Device: /dev/sda, starting scheduled Short Self-Test. Nov 10 06:33:03 subversion smartd[30248]: Device: /dev/sdb, starting scheduled Short Self-Test. As you can see it has no problem doing the short tests! I am going to experiment. I have taken the check intervals down from 2hrs to 65mins and also I am going to switch over the long and short tests just by switching the L and S to see if it is for some reason only dealing with the first one.. |
From: Christian F. <Chr...@t-...> - 2008-11-14 10:06:53
|
Jon Hardcastle wrote: > [...] > > As you can see it has no problem doing the short tests! > > I am going to experiment. I have taken the check intervals down from > 2hrs to 65mins and also I am going to switch over the long and short > tests just by switching the L and S to see if it is for some reason > only dealing with the first one.. > Reduce the check interval to 60min or less. Otherwise, the check time may be not within the hour specified in the regexp. Then the regexp does not match and the test is skipped (see smartd.conf man page). The match algorithm was changed after the 5.38 release. With current CVS version, scheduled tests should work also with intervals greater than 60min. Even tests scheduled during downtime are run after next boot if the new --savestates option is used. Cheers, Christian |
From: Jon H. <jd_...@ya...> - 2008-11-14 10:24:43
|
--- On Fri, 14/11/08, Christian Franke <Chr...@t-...> wrote: > From: Christian Franke <Chr...@t-...> > Subject: Re: [smartmontools-support] Anyone tell me.. > To: Jon@eHardcastle.com > Cc: sma...@li..., "Bruce Allen" <ba...@gr...> > Date: Friday, 14 November, 2008, 10:05 AM > Jon Hardcastle wrote: > > [...] > > > > As you can see it has no problem doing the short > tests! > > > > I am going to experiment. I have taken the check > intervals down from > > 2hrs to 65mins and also I am going to switch over the > long and short > > tests just by switching the L and S to see if it is > for some reason > > only dealing with the first one.. > > > > Reduce the check interval to 60min or less. Otherwise, the > check time > may be not within the hour specified in the regexp. Then > the regexp does > not match and the test is skipped (see smartd.conf man > page). > > The match algorithm was changed after the 5.38 release. > With current CVS > version, scheduled tests should work also with intervals > greater than > 60min. Even tests scheduled during downtime are run after > next boot if > the new --savestates option is used. > > Cheers, > Christian That sounds excellent! Unfortunately upgrading the system at work isn't easy. But i'll get that home definitely. Another issue I have been having for ages is on my server at home the drives go to sleep after the machine hasn't been used for a while when the scheduled check comes along it spins the disc up but for some reason the drive doesn't spin up quick enough and i get a whole raft of emails telling me the drive isn't smart capable but it seems to actually do the test? i just loads of stuff dumped into messages/dmesg? |