System started crashing hard after zpool reconfigure and OI upgrade

Discussion:

Peter Wood

2013-03-20 15:34:36 UTC

I have two identical Supermicro boxes with 32GB ram. Hardware details at
the end of the message.

They were running OI 151.a.5 for months. The zpool configuration was one
storage zpool with 3 vdevs of 8 disks in RAIDZ2.

The OI installation is absolutely clean. Just next-next-next until done.
All I do is configure the network after install. I don't install or enable
any other services.

Then I added more disks and rebuild the systems with OI 151.a.7 and this
time configured the zpool with 6 vdevs of 5 disks in RAIDZ.

The systems started crashing really bad. They just disappear from the
network, black and unresponsive console, no error lights but no activity
indication either. The only way out is to power cycle the system.

There is no pattern in the crashes. It may crash in 2 days in may crash in
2 hours.

I upgraded the memory on both systems to 128GB at no avail. This is the max
memory they can take.

In summary all I did is upgrade to OI 151.a.7 and reconfigured zpool.

Any idea what could be the problem.

Thank you

-- Peter

Supermicro X9DRH-iF
Xeon E5-2620 @ 2.0 GHz 6-Core
LSI SAS9211-8i HBA
32x 3TB Hitachi HUS723030ALS640, SAS, 7.2K

Michael Schuster

2013-03-20 15:38:39 UTC

Permalink

Peter,

sorry if this is so obvious that you didn't mention it: Have you checked
/var/adm/messages and other diagnostic tool output?

regards
Michael

Post by Peter Wood
I have two identical Supermicro boxes with 32GB ram. Hardware details at
the end of the message.
They were running OI 151.a.5 for months. The zpool configuration was one
storage zpool with 3 vdevs of 8 disks in RAIDZ2.
The OI installation is absolutely clean. Just next-next-next until done.
All I do is configure the network after install. I don't install or enable
any other services.
Then I added more disks and rebuild the systems with OI 151.a.7 and this
time configured the zpool with 6 vdevs of 5 disks in RAIDZ.
The systems started crashing really bad. They just disappear from the
network, black and unresponsive console, no error lights but no activity
indication either. The only way out is to power cycle the system.
There is no pattern in the crashes. It may crash in 2 days in may crash in
2 hours.
I upgraded the memory on both systems to 128GB at no avail. This is the
max memory they can take.
In summary all I did is upgrade to OI 151.a.7 and reconfigured zpool.
Any idea what could be the problem.
Thank you
-- Peter
Supermicro X9DRH-iF
LSI SAS9211-8i HBA
32x 3TB Hitachi HUS723030ALS640, SAS, 7.2K
_______________________________________________
zfs-discuss mailing list
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

--
Michael Schuster
http://recursiveramblings.wordpress.com/

Will Murnane

2013-03-20 15:40:33 UTC

Permalink

Does the Supermicro IPMI show anything when it crashes? Does anything show
up in event logs in the BIOS, or in system logs under OI?

Peter Wood

2013-03-20 15:50:40 UTC

Permalink

I'm sorry. I should have mentioned it that I can't find any errors in the
logs. The last entry in /var/adm/messages is that I removed the keyboard
after the last reboot and then it shows the new boot up messages when I
boot up the system after the crash. The BIOS log is empty. I'm not sure how
to check the IPMI but IPMI is not configured and I'm not using it.

Just another observation - the crashes are more intense the more data the
system serves (NFS).

I'm looking into FRMW upgrades for the LSI now.

Post by Will Murnane
Does the Supermicro IPMI show anything when it crashes? Does anything
show up in event logs in the BIOS, or in system logs under OI?

Post by Peter Wood
I have two identical Supermicro boxes with 32GB ram. Hardware details at
the end of the message.
They were running OI 151.a.5 for months. The zpool configuration was one
storage zpool with 3 vdevs of 8 disks in RAIDZ2.
The OI installation is absolutely clean. Just next-next-next until done.
All I do is configure the network after install. I don't install or enable
any other services.
Then I added more disks and rebuild the systems with OI 151.a.7 and this
time configured the zpool with 6 vdevs of 5 disks in RAIDZ.
The systems started crashing really bad. They just disappear from the
network, black and unresponsive console, no error lights but no activity
indication either. The only way out is to power cycle the system.
There is no pattern in the crashes. It may crash in 2 days in may crash
in 2 hours.
I upgraded the memory on both systems to 128GB at no avail. This is the
max memory they can take.
In summary all I did is upgrade to OI 151.a.7 and reconfigured zpool.
Any idea what could be the problem.
Thank you
-- Peter
Supermicro X9DRH-iF
LSI SAS9211-8i HBA
32x 3TB Hitachi HUS723030ALS640, SAS, 7.2K
_______________________________________________
zfs-discuss mailing list
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Michael Schuster

2013-03-20 15:53:03 UTC

Permalink

How about crash dumps?

michael

Post by Peter Wood
I'm sorry. I should have mentioned it that I can't find any errors in the
logs. The last entry in /var/adm/messages is that I removed the keyboard
after the last reboot and then it shows the new boot up messages when I
boot up the system after the crash. The BIOS log is empty. I'm not sure how
to check the IPMI but IPMI is not configured and I'm not using it.
Just another observation - the crashes are more intense the more data the
system serves (NFS).
I'm looking into FRMW upgrades for the LSI now.

Post by Will Murnane
Does the Supermicro IPMI show anything when it crashes? Does anything
show up in event logs in the BIOS, or in system logs under OI?

Post by Peter Wood
I have two identical Supermicro boxes with 32GB ram. Hardware details at
the end of the message.
They were running OI 151.a.5 for months. The zpool configuration was one
storage zpool with 3 vdevs of 8 disks in RAIDZ2.
The OI installation is absolutely clean. Just next-next-next until done.
All I do is configure the network after install. I don't install or enable
any other services.
Then I added more disks and rebuild the systems with OI 151.a.7 and this
time configured the zpool with 6 vdevs of 5 disks in RAIDZ.
The systems started crashing really bad. They just disappear from the
network, black and unresponsive console, no error lights but no activity
indication either. The only way out is to power cycle the system.
There is no pattern in the crashes. It may crash in 2 days in may crash
in 2 hours.
I upgraded the memory on both systems to 128GB at no avail. This is the
max memory they can take.
In summary all I did is upgrade to OI 151.a.7 and reconfigured zpool.
Any idea what could be the problem.
Thank you
-- Peter
Supermicro X9DRH-iF
LSI SAS9211-8i HBA
32x 3TB Hitachi HUS723030ALS640, SAS, 7.2K
_______________________________________________
zfs-discuss mailing list
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

_______________________________________________
zfs-discuss mailing list
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

--
Michael Schuster
http://recursiveramblings.wordpress.com/

Peter Wood

2013-03-20 16:15:27 UTC

Permalink

I'm going to need some help with the crash dumps. I'm not very familiar
with Solaris.

Do I have to enable something to get the crash dumps? Where should I look
for them?

Thanks for the help.

Post by Michael Schuster
How about crash dumps?
michael

Post by Will Murnane
Does the Supermicro IPMI show anything when it crashes? Does anything
show up in event logs in the BIOS, or in system logs under OI?

Post by Peter Wood
I have two identical Supermicro boxes with 32GB ram. Hardware details
at the end of the message.
They were running OI 151.a.5 for months. The zpool configuration was
one storage zpool with 3 vdevs of 8 disks in RAIDZ2.
The OI installation is absolutely clean. Just next-next-next until
done. All I do is configure the network after install. I don't install or
enable any other services.
Then I added more disks and rebuild the systems with OI 151.a.7 and
this time configured the zpool with 6 vdevs of 5 disks in RAIDZ.
The systems started crashing really bad. They just disappear from the
network, black and unresponsive console, no error lights but no activity
indication either. The only way out is to power cycle the system.
There is no pattern in the crashes. It may crash in 2 days in may crash
in 2 hours.
I upgraded the memory on both systems to 128GB at no avail. This is the
max memory they can take.
In summary all I did is upgrade to OI 151.a.7 and reconfigured zpool.
Any idea what could be the problem.
Thank you
-- Peter
Supermicro X9DRH-iF
LSI SAS9211-8i HBA
32x 3TB Hitachi HUS723030ALS640, SAS, 7.2K
_______________________________________________
zfs-discuss mailing list
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

_______________________________________________
zfs-discuss mailing list
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

--
Michael Schuster
http://recursiveramblings.wordpress.com/

Jim Klimov

2013-03-20 18:29:11 UTC

Permalink

Post by Peter Wood
I'm going to need some help with the crash dumps. I'm not very familiar
with Solaris.
Do I have to enable something to get the crash dumps? Where should I
look for them?

Typically the kernel crash dumps are created as a result of kernel
panic; also they may be forced by administrative actions like NMI.
They require you to configure a dump volume of sufficient size (see
dumpadm) and a /var/crash which may be a dataset on a large enough
pool - after the reboot the dump data will be migrated there.

To "help" with the hangs you can try the BIOS watchdog (which would
require a bmc driver, one which is known from OpenSolaris is alas
not opensourced and not redistributable), or with a software deadman
timer:

http://www.cuddletech.com/blog/pivot/entry.php?id=1044

http://wiki.illumos.org/display/illumos/System+Hangs

Also, if you configure "crash dump on NMI" and set up your IPMI card,
then you can likely gain remote access to both the server console
("physical" and/or serial) and may be able to trigger the NMI, too.

HTH,
//Jim

Post by Peter Wood
Thanks for the help.
On Wed, Mar 20, 2013 at 8:53 AM, Michael Schuster
How about crash dumps?
michael
I'm sorry. I should have mentioned it that I can't find any
errors in the logs. The last entry in /var/adm/messages is that
I removed the keyboard after the last reboot and then it shows
the new boot up messages when I boot up the system after the
crash. The BIOS log is empty. I'm not sure how to check the IPMI
but IPMI is not configured and I'm not using it.
Just another observation - the crashes are more intense the more
data the system serves (NFS).
I'm looking into FRMW upgrades for the LSI now.
On Wed, Mar 20, 2013 at 8:40 AM, Will Murnane
Does the Supermicro IPMI show anything when it crashes?
Does anything show up in event logs in the BIOS, or in
system logs under OI?
On Wed, Mar 20, 2013 at 11:34 AM, Peter Wood
I have two identical Supermicro boxes with 32GB ram.
Hardware details at the end of the message.
They were running OI 151.a.5 for months. The zpool
configuration was one storage zpool with 3 vdevs of 8
disks in RAIDZ2.
The OI installation is absolutely clean. Just
next-next-next until done. All I do is configure the
network after install. I don't install or enable any
other services.
Then I added more disks and rebuild the systems with OI
151.a.7 and this time configured the zpool with 6 vdevs
of 5 disks in RAIDZ.
The systems started crashing really bad. They
just disappear from the network, black and unresponsive
console, no error lights but no activity indication
either. The only way out is to power cycle the system.
There is no pattern in the crashes. It may crash in 2
days in may crash in 2 hours.
I upgraded the memory on both systems to 128GB at no
avail. This is the max memory they can take.
In summary all I did is upgrade to OI 151.a.7 and
reconfigured zpool.
Any idea what could be the problem.
Thank you
-- Peter
Supermicro X9DRH-iF
LSI SAS9211-8i HBA
32x 3TB Hitachi HUS723030ALS640, SAS, 7.2K
_______________________________________________
zfs-discuss mailing list
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
_______________________________________________
zfs-discuss mailing list
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
--
Michael Schuster
http://recursiveramblings.wordpress.com/
_______________________________________________
zfs-discuss mailing list
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Peter Wood

2013-03-20 19:38:27 UTC

Permalink

Hi Jim,

Thanks for the pointers. I'll definitely look into this.

--
Peter Blajev
IT Manager, TAAZ Inc.
Office: 858-597-0512 x125

Post by Jim Klimov

Post by Peter Wood
I'm going to need some help with the crash dumps. I'm not very familiar
with Solaris.
Do I have to enable something to get the crash dumps? Where should I
look for them?

Typically the kernel crash dumps are created as a result of kernel
panic; also they may be forced by administrative actions like NMI.
They require you to configure a dump volume of sufficient size (see
dumpadm) and a /var/crash which may be a dataset on a large enough
pool - after the reboot the dump data will be migrated there.
To "help" with the hangs you can try the BIOS watchdog (which would
require a bmc driver, one which is known from OpenSolaris is alas
not opensourced and not redistributable), or with a software deadman
http://www.cuddletech.com/**blog/pivot/entry.php?id=1044<http://www.cuddletech.com/blog/pivot/entry.php?id=1044>
http://wiki.illumos.org/**display/illumos/System+Hangs<http://wiki.illumos.org/display/illumos/System+Hangs>
Also, if you configure "crash dump on NMI" and set up your IPMI card,
then you can likely gain remote access to both the server console
("physical" and/or serial) and may be able to trigger the NMI, too.
HTH,
//Jim

Post by Peter Wood
Thanks for the help.
On Wed, Mar 20, 2013 at 8:53 AM, Michael Schuster
How about crash dumps?
michael
I'm sorry. I should have mentioned it that I can't find any
errors in the logs. The last entry in /var/adm/messages is that
I removed the keyboard after the last reboot and then it shows
the new boot up messages when I boot up the system after the
crash. The BIOS log is empty. I'm not sure how to check the IPMI
but IPMI is not configured and I'm not using it.
Just another observation - the crashes are more intense the more
data the system serves (NFS).
I'm looking into FRMW upgrades for the LSI now.
On Wed, Mar 20, 2013 at 8:40 AM, Will Murnane
Does the Supermicro IPMI show anything when it crashes?
Does anything show up in event logs in the BIOS, or in
system logs under OI?
On Wed, Mar 20, 2013 at 11:34 AM, Peter Wood
I have two identical Supermicro boxes with 32GB ram.
Hardware details at the end of the message.
They were running OI 151.a.5 for months. The zpool
configuration was one storage zpool with 3 vdevs of 8
disks in RAIDZ2.
The OI installation is absolutely clean. Just
next-next-next until done. All I do is configure the
network after install. I don't install or enable any
other services.
Then I added more disks and rebuild the systems with OI
151.a.7 and this time configured the zpool with 6 vdevs
of 5 disks in RAIDZ.
The systems started crashing really bad. They
just disappear from the network, black and unresponsive
console, no error lights but no activity indication
either. The only way out is to power cycle the system.
There is no pattern in the crashes. It may crash in 2
days in may crash in 2 hours.
I upgraded the memory on both systems to 128GB at no
avail. This is the max memory they can take.
In summary all I did is upgrade to OI 151.a.7 and
reconfigured zpool.
Any idea what could be the problem.
Thank you
-- Peter
Supermicro X9DRH-iF
LSI SAS9211-8i HBA
32x 3TB Hitachi HUS723030ALS640, SAS, 7.2K
______________________________**_________________
zfs-discuss mailing list
http://mail.opensolaris.org/**
mailman/listinfo/zfs-discuss<http://mail.opensolaris.org/mailman/listinfo/zfs-discuss>
______________________________**_________________
zfs-discuss mailing list
http://mail.opensolaris.org/**mailman/listinfo/zfs-discuss<http://mail.opensolaris.org/mailman/listinfo/zfs-discuss>
--
Michael Schuster
http://recursiveramblings.**wordpress.com/<http://recursiveramblings.wordpress.com/>
______________________________**_________________
zfs-discuss mailing list
http://mail.opensolaris.org/**mailman/listinfo/zfs-discuss<http://mail.opensolaris.org/mailman/listinfo/zfs-discuss>

______________________________**_________________
zfs-discuss mailing list
http://mail.opensolaris.org/**mailman/listinfo/zfs-discuss<http://mail.opensolaris.org/mailman/listinfo/zfs-discuss>

Jens Elkner

2013-03-20 20:08:07 UTC

Permalink

Post by Peter Wood
I'm sorry. I should have mentioned it that I can't find any errors in the
logs. The last entry in /var/adm/messages is that I removed the keyboard
after the last reboot and then it shows the new boot up messages when I boot
up the system after the crash. The BIOS log is empty. I'm not sure how to
check the IPMI but IPMI is not configured and I'm not using it.

You definitely should! Plugin a cable into the dedicated network port
and configure it (easiest way for you is probably to jump into the BIOS
and assign the appropriate IP address etc.). Than, for a quick look,
point your browser to the given IP port 80 (default login is
ADMIN/ADMIN). Also you may now configure some other details
(accounts/passwords/roles).

To track the problem, either write a script, which polls the parameters
in question periodically or just install the latest ipmiViewer and use
this to monitor your sensors ad hoc.
see ftp://ftp.supermicro.com/utility/IPMIView/

Post by Peter Wood
Just another observation - the crashes are more intense the more data the
system serves (NFS).
I'm looking into FRMW upgrades for the LSI now.

Latest LSI FW should be P15, for this MB type 217 (2.17), MB-BIOS C28 (1.0b).
However, I doubt, that your problem has anything to do with the
SAS-ctrl or OI or ZFS.

My guess is, that either your MB is broken (we had an X9DRH-iF, which
instantly "disappeared" as soon as it got some real load) or you have
a heat problem (watch you cpu temp e.g. via ipmiviewer). With 2GHz
that's not very likely, but worth a try (socket placement on this board
is not really smart IMHO).

To test quickly
- disable all addtional, unneeded service in OI, which may put some
load on the machine (like NFS service, http and bla) and perhaps
even export unneeded pools (just to be sure)
- fire up your ipmiviewer and look at the sensors (set update to
10s) or refresh manually often
- start 'openssl speed -multi 32' and keep watching your cpu temp
sensors (with 2GHz I guess it takes ~ 12min)

I guess, your machine "disappears" before the CPUs getting really hot
(broken MB). If CPUs switch off (usually first CPU2 and a little bit
later CPU1) you have a cooling problem. If nothing happens, well, than
it could be an OI or ZFS problem ;-)

Have fun,
jel.

--
Otto-von-Guericke University http://www.cs.uni-magdeburg.de/
Department of Computer Science Geb. 29 R 027, Universitaetsplatz 2
39106 Magdeburg, Germany Tel: +49 391 67 52768

Peter Wood

2013-03-20 20:45:53 UTC

Permalink

Great write up Jens.

The chance of two MB to be broken is probably low but overheating is a very
good point. It was on my to-do list to setup IPMI and seems that now is the
best time to do it.

Thanks

Post by Peter Wood

Post by Peter Wood
I'm sorry. I should have mentioned it that I can't find any errors in

the

Post by Peter Wood
logs. The last entry in /var/adm/messages is that I removed the

keyboard

Post by Peter Wood
after the last reboot and then it shows the new boot up messages when

I boot

Post by Peter Wood
up the system after the crash. The BIOS log is empty. I'm not sure

how to

Post by Peter Wood
check the IPMI but IPMI is not configured and I'm not using it.

You definitely should! Plugin a cable into the dedicated network port
and configure it (easiest way for you is probably to jump into the BIOS
and assign the appropriate IP address etc.). Than, for a quick look,
point your browser to the given IP port 80 (default login is
ADMIN/ADMIN). Also you may now configure some other details
(accounts/passwords/roles).
To track the problem, either write a script, which polls the parameters
in question periodically or just install the latest ipmiViewer and use
this to monitor your sensors ad hoc.
see ftp://ftp.supermicro.com/utility/IPMIView/

Post by Peter Wood
Just another observation - the crashes are more intense the more data

the

Post by Peter Wood
system serves (NFS).
I'm looking into FRMW upgrades for the LSI now.

Latest LSI FW should be P15, for this MB type 217 (2.17), MB-BIOS C28 (1.0b).
However, I doubt, that your problem has anything to do with the
SAS-ctrl or OI or ZFS.
My guess is, that either your MB is broken (we had an X9DRH-iF, which
instantly "disappeared" as soon as it got some real load) or you have
a heat problem (watch you cpu temp e.g. via ipmiviewer). With 2GHz
that's not very likely, but worth a try (socket placement on this board
is not really smart IMHO).
To test quickly
- disable all addtional, unneeded service in OI, which may put some
load on the machine (like NFS service, http and bla) and perhaps
even export unneeded pools (just to be sure)
- fire up your ipmiviewer and look at the sensors (set update to
10s) or refresh manually often
- start 'openssl speed -multi 32' and keep watching your cpu temp
sensors (with 2GHz I guess it takes ~ 12min)
I guess, your machine "disappears" before the CPUs getting really hot
(broken MB). If CPUs switch off (usually first CPU2 and a little bit
later CPU1) you have a cooling problem. If nothing happens, well, than
it could be an OI or ZFS problem ;-)
Have fun,
jel.
--
Otto-von-Guericke University http://www.cs.uni-magdeburg.de/
Department of Computer Science Geb. 29 R 027, Universitaetsplatz 2
39106 Magdeburg, Germany Tel: +49 391 67 52768
_______________________________________________
zfs-discuss mailing list
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Peter Wood

2013-03-20 19:42:22 UTC

Permalink

No problem Trey. Anything will help.

Yes, I did a clean install overwriting the old OS.

Just to make sure, you actually did an overwrite reinstall with OI151a7
rather than upgrading the existing OS images? If you did a pkg
image-update, you should be able to boot back into the oi151a5 image from
grub. Apologies in advance if I'm stating the obvious.
-- Trey
I have two identical Supermicro boxes with 32GB ram. Hardware details
at the end of the message.
They were running OI 151.a.5 for months. The zpool configuration was one
storage zpool with 3 vdevs of 8 disks in RAIDZ2.
The OI installation is absolutely clean. Just next-next-next until done.
All I do is configure the network after install. I don't install or enable
any other services.
Then I added more disks and rebuild the systems with OI 151.a.7 and this
time configured the zpool with 6 vdevs of 5 disks in RAIDZ.
The systems started crashing really bad. They just disappear from the
network, black and unresponsive console, no error lights but no activity
indication either. The only way out is to power cycle the system.
There is no pattern in the crashes. It may crash in 2 days in may crash
in 2 hours.
I upgraded the memory on both systems to 128GB at no avail. This is the
max memory they can take.
In summary all I did is upgrade to OI 151.a.7 and reconfigured zpool.
Any idea what could be the problem.
Thank you
-- Peter
Supermicro X9DRH-iF
LSI SAS9211-8i HBA
32x 3TB Hitachi HUS723030ALS640, SAS, 7.2K
_______________________________________________
zfs-discuss mailing list
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Peter Wood

2013-03-20 21:34:36 UTC

Permalink

I can reproduce the problem. I can crash the system.

Here are the steps I did (some steps may not be needed but I haven't tested
it):

- Clean install of OI 151.a.7 on Supermicro hardware described above (32GB
RAM though, not the 128GB)

- Create 1 zpool, 6 raidz vdevs with 5 drives each

- NFS export a dataset
zfs set sharenfs="rw=@10.20.1/24" vol01/htmlspace

- Create zfs child dataset
zfs create vol01/htmlspace/A

$ zfs get -H sharenfs vol01/htmlspace/A
vol01/htmlspace/A sharenfs rw=@10.20.1/24 inherited from
vol01/htmlspace

- Stop NFS shearing for the child dataset

zfs set sharenfs=off vol01/htmlspace/A

The crash is instant after the sharenfs=off command.

I thought it was coincident so after reboot I tried it on another dataset.
Instant crash again. I get my prompt back but that's it. The system is gone
after that.

The NFS exported file systems are not accessed by any system on the
network. They are not in use. That's why I wanted to stop exporting them.
And, even if they were in use this should now crash the system, right?

I can't try the other box because it is heavy in production. At least not
until later tonight.

I thought I'll collect some advice to make each crash as useful as possible.

Any pointers are appreciated.

Thanks,

-- Peter