From: Steve W. <st...@ca...> - 2004-06-04 03:34:36
|
Hi there! We have some old quantum bigfoot drives that we are going to have to use. We want to set up smartd to email us when they start failing. I'd like to be able to properly test this before deploying them, so I took to one of these drives with a hammer while it was running, and checking smartctl to see if it noticed. We got kernel errors, but no smart errors reported. Unfortunately, it seems that inducing errors with a hammer has two states; 1. drive doesn't realy care 2. drive is totally dead Has anyone any suggestions as to how we could induce smart errors without actually destroying the disk? Thanks! |
From: Bruce A. <ba...@gr...> - 2004-06-04 16:10:15
|
Hi Steve, I can think of two ways to do this. First, you can try to reset the SMART failure threshold values. But you'll need to write code to do this, and hope that the drives implement the undocumented SMART WRITE THRESHOLDS command. Alternatively, I've been told you can put the drives on a shaker table and turn up the amplitude. But once the drives have failing SMART status, they are likely to die and so shouldn't be used anymore. Good luck! Cheers, Bruce On Fri, 4 Jun 2004, Steve Wray wrote: > Hi there! > > We have some old quantum bigfoot drives that we are going to have to > use. We want to set up smartd to email us when they start failing. > > I'd like to be able to properly test this before deploying them, so I > took to one of these drives with a hammer while it was running, and > checking smartctl to see if it noticed. > > We got kernel errors, but no smart errors reported. > > Unfortunately, it seems that inducing errors with a hammer has two > states; > > 1. drive doesn't realy care > 2. drive is totally dead > > Has anyone any suggestions as to how we could induce smart errors > without actually destroying the disk? > > Thanks! > > > ------------------------------------------------------- > This SF.Net email is sponsored by the new InstallShield X. > From Windows to Linux, servers to mobile, InstallShield X is the one > installation-authoring solution that does it all. Learn more and > evaluate today! http://www.installshield.com/Dev2Dev/0504 > _______________________________________________ > Smartmontools-support mailing list > Sma...@li... > https://lists.sourceforge.net/lists/listinfo/smartmontools-support > > |
From: Steve W. <st...@ca...> - 2004-06-05 08:00:27
|
> Bruce Allen wrote: > Hi Steve, > > I can think of two ways to do this. First, you can try to > reset the SMART failure threshold values. But you'll need to write code to do > this, and hope that the drives implement the undocumented SMART WRITE THRESHOLDS > command. That could save me some trouble, I'll look into it > Alternatively, I've been told you can put the drives on a > shaker table and turn up the amplitude. > But once the drives have failing SMART status, > they are likely to die and so shouldn't be used anymore. Absolutely. These are 'sacrificial drives' to prove the notification system. The boxes are going to be distributed around the country so ideally we want to have plenty of advance warning of a failing drive. I had been worrying that the old bigfoot didn't support smart very well; it doesn't do any logging. > Good luck! > > Cheers, > Bruce > > On Fri, 4 Jun 2004, Steve Wray wrote: > > > Hi there! > > > > We have some old quantum bigfoot drives that we are going > to have to > > use. We want to set up smartd to email us when they start failing. > > > > I'd like to be able to properly test this before deploying > them, so I > > took to one of these drives with a hammer while it was running, and > > checking smartctl to see if it noticed. > > > > We got kernel errors, but no smart errors reported. > > > > Unfortunately, it seems that inducing errors with a hammer has two > > states; > > > > 1. drive doesn't realy care > > 2. drive is totally dead > > > > Has anyone any suggestions as to how we could induce smart errors > > without actually destroying the disk? > > > > Thanks! > > > > > > ------------------------------------------------------- > > This SF.Net email is sponsored by the new InstallShield X. > > From Windows to Linux, servers to mobile, InstallShield X is the one > > installation-authoring solution that does it all. Learn more and > > evaluate today! http://www.installshield.com/Dev2Dev/0504 > > _______________________________________________ > > Smartmontools-support mailing list > > Sma...@li... > > https://lists.sourceforge.net/lists/listinfo/smartmontools-support > > > > > > > > ------------------------------------------------------- > This SF.Net email is sponsored by the new InstallShield X. > From Windows to Linux, servers to mobile, InstallShield X is the one > installation-authoring solution that does it all. Learn more and > evaluate today! http://www.installshield.com/Dev2Dev/0504 > _______________________________________________ > Smartmontools-support mailing list > Sma...@li... > https://lists.sourceforge.net/lists/listinfo/smartmontools-support > |
From: Hugo M. C. <hc...@ma...> - 2004-06-05 19:00:26
|
Dear all, I would like to reduce the number of messages produced in /var/log/messages. A particular example of the messages that I would like to diminish are the following: Jun 5 06:49:08 migr003d smartd[25008]: Device: /dev/sda [3ware_disk_04], SMART Prefailure Attribute: 7 Seek_Error_Rate changed from 200 to 100 Jun 5 07:19:07 migr003d smartd[25008]: Device: /dev/sda [3ware_disk_02], SMART Prefailure Attribute: 7 Seek_Error_Rate changed from 200 to 100 Jun 5 07:19:09 migr003d smartd[25008]: Device: /dev/sda [3ware_disk_06], SMART Prefailure Attribute: 7 Seek_Error_Rate changed from 200 to 100 Jun 5 07:19:10 migr003d smartd[25008]: Device: /dev/sdb [3ware_disk_02], SMART Prefailure Attribute: 7 Seek_Error_Rate changed from 200 to 100 Jun 5 07:49:08 migr003d smartd[25008]: Device: /dev/sda [3ware_disk_06], SMART Prefailure Attribute: 1 Raw_Read_Error_Rate changed from 200 to 199 Jun 5 08:19:26 migr003d smartd[25008]: Device: /dev/sda [3ware_disk_06], How can I specify in the smartd.conf file that I don't want these messages to appear in the /var/log/messages ? I saw that in the man page it talked about the -I directive that would ignore certain attributes ... but will this ignore the attributes or will just not report it in /var/log/messages ?? Is there any way to specify that I want to ignore all the attributes instead of one by one ... something like -I all ??? Thank you, Hugo Caçote |
From: Bruce A. <ba...@gr...> - 2004-06-06 02:33:52
|
Quoting from the smartd manual page: If you are not sure which Directives to use, I suggest experimenting for a few minutes with smartctl to see what SMART functionality your disk(s) support(s). If you do not like voluminous syslog messages, a good choice of smartd configuration file Directives might be: -H -l selftest -l error -f. If you want more frequent information, use: -a. On Sat, 5 Jun 2004, Hugo Monteiro Cacote wrote: > Dear all, >=20 > I would like to reduce the number of messages produced in=20 > /var/log/messages. A particular example of the messages that I would like= =20 > to diminish are the following: >=20 > Jun 5 06:49:08 migr003d smartd[25008]: Device: /dev/sda [3ware_disk_04],= =20 > SMART Prefailure Attribute: 7 Seek_Error_Rate changed from 200 to 100=20 > Jun 5 07:19:07 migr003d smartd[25008]: Device: /dev/sda [3ware_disk_02],= =20 > SMART Prefailure Attribute: 7 Seek_Error_Rate changed from 200 to 100=20 > Jun 5 07:19:09 migr003d smartd[25008]: Device: /dev/sda [3ware_disk_06],= =20 > SMART Prefailure Attribute: 7 Seek_Error_Rate changed from 200 to 100=20 > Jun 5 07:19:10 migr003d smartd[25008]: Device: /dev/sdb [3ware_disk_02],= =20 > SMART Prefailure Attribute: 7 Seek_Error_Rate changed from 200 to 100=20 > Jun 5 07:49:08 migr003d smartd[25008]: Device: /dev/sda [3ware_disk_06],= =20 > SMART Prefailure Attribute: 1 Raw_Read_Error_Rate changed from 200 to 199= =20 > Jun 5 08:19:26 migr003d smartd[25008]: Device: /dev/sda [3ware_disk_06],= =20 >=20 > How can I specify in the smartd.conf file that I don't want these message= s=20 > to appear in the /var/log/messages ? I saw that in the man page it talked= =20 > about the -I directive that would ignore certain attributes ... but will= =20 > this ignore the attributes or will just not report it in=20 > /var/log/messages ?? Is there any way to specify that I want to ignore= =20 > all the attributes instead of one by one ... something like -I all ??? >=20 > Thank you, > Hugo Ca=E7ote >=20 >=20 >=20 >=20 > ------------------------------------------------------- > This SF.Net email is sponsored by the new InstallShield X. > From Windows to Linux, servers to mobile, InstallShield X is the one > installation-authoring solution that does it all. Learn more and > evaluate today! http://www.installshield.com/Dev2Dev/0504 > _______________________________________________ > Smartmontools-support mailing list > Sma...@li... > https://lists.sourceforge.net/lists/listinfo/smartmontools-support >=20 >=20 |