Just a poll to see who attempts to disconnect/unmount every possible client (server) before attempting a DataONTAP upgrade and who types download; reboot.
Does anyone leave their Oracle databases mounted via NFS over UDP up the whole time?
We stop the databases, but leave the filesystem mounted. We've had the filer panic and reboot with everything up and running with no problems, but it's not something we'd do on purpose.
My .02, Jeff Mery, MCP National Instruments
------------------------------------------------------------------------- "Allow me to extol the virtues of the Net Fairy, and of all the fantastic dorks that make the nice packets go from here to there. Amen." TB - Penny Arcade -------------------------------------------------------------------------
Ambrose_Earle@shamrockfoods.com Sent by: owner-toasters@mathworks.com 10/22/2003 05:50 PM
To: toasters@mathworks.com cc: Subject: Filer upgrades
Just a poll to see who attempts to disconnect/unmount every possible client (server) before attempting a DataONTAP upgrade and who types download; reboot.
Does anyone leave their Oracle databases mounted via NFS over UDP up the whole time?
On Wed, 22 Oct 2003 Ambrose_Earle@shamrockfoods.com wrote:
Just a poll to see who attempts to disconnect/unmount every possible client (server) before attempting a DataONTAP upgrade and who types download; reboot.
Does anyone leave their Oracle databases mounted via NFS over UDP up the whole time?
Heavens, we always sync Oracle. On production DBs where we can't afford a full shutdown/restart cycle, I think the DBA puts it in hot backup mode (even when I'm about to replace a disk, just on the off-chance that something goes haywire) and the reboot cycle is quick enough that things just come right back. When we can afford the downtime, however, it is preferable to shut Oracle down properly, reboot, then bring it back up.
But for normal NFS clients on filers not serving databases we just bounce 'em. Y'know, the once a year or so that it's necessary:
4:37pm up 234 days, 1:44 2026468359 NFS ops, 0 CIFS ops, 0 HTTP ops, 0 DAFS ops 4:37pm up 255 days, 20:47 1725238310 NFS ops, 18226083 CIFS ops, 0 HTTP ops, 0 DAFS ops 4:38pm up 58 days, 3:12 31217232 NFS ops, 0 CIFS ops, 0 HTTP ops, 0 DAFS ops
(Okay, on that last one, let me tell you about the fun we had when the UPS technician came out and the transfer switch... didn't...)
:-)
Hey, what's that command to have ONTAP print out uptime stats? availsomething? I'm curious to see what our stats have been over the last three years.
-- Chris Lamb, Unix Guy SelectMetrics, Inc. 503-241-1469 x247 skeezics@selectmetrics.com
Chris Lamb writes:
Hey, what's that command to have ONTAP print out uptime stats? availsomething? I'm curious to see what our stats have been over the last three years.
"availtime", or "availtime full" for extra verbosity. One of those commands sneakily omitted from the list given by "?".
Chris Thompson Email: cet1@cam.ac.uk
On Wed, 22 Oct 2003 Ambrose_Earle@shamrockfoods.com wrote:
Just a poll to see who attempts to disconnect/unmount every possible client (server) before attempting a DataONTAP upgrade and who types download; reboot.
Does anyone leave their Oracle databases mounted via NFS over UDP up the whole time?
For our databases, we usually shut them down before doing anything on the filers, if we can. A couple weeks ago, a cluster parter rebooted then took down its partner on its way up; the DBAs never even noticed. When they asked about the cluster partner we told them that both had gone down and they asked if we were serious. When we didn't laugh, they rushed to check their logs to make sure everything was ok (and everything was ok).
By and large, if we have individual NFS clients attached during a reboot, they can the 30 seconds it takes for the filer to reboot. We have had isolated instances where older SunOS systems (and maybe our Solaris 2.6 FlexLM license servers) have a problem with automounted NFS mounts, so those are usually shutdown in the event we have some planned downtime. Other than that, most people just blame it on the network and keep on going.
Geoff
Geoff Hardin ghardin@dalsemi.com writes [...]
By and large, if we have individual NFS clients attached during a reboot, they can the 30 seconds it takes for the filer to reboot. We have had isolated instances where older SunOS systems (and maybe our Solaris 2.6 FlexLM license servers) have a problem with automounted NFS mounts, so those are usually shutdown in the event we have some planned downtime. Other than that, most people just blame it on the network and keep on going.
We have had a problem with the automounter even in recent versions of Solaris (including one instance with Solaris 9 so far). It seems to be triggered by the automounter deciding to try unmounting an NFS filing system while the server is unresponsive. The result is that that all accesses to the mounted filing system stall indefinitely (but interruptibly) even after the server is back.
This is low-frequency for any particular server reboot, but still sufficiently common that we try to schedule things so that we can follow up the server reboot by a client reboot if it strikes.
Chris Thompson Email: cet1@cam.ac.uk
On Thu, Oct 23, 2003 at 12:08:03PM +0100, Chris Thompson wrote:
We have had a problem with the automounter even in recent versions of Solaris (including one instance with Solaris 9 so far). It seems to be triggered by the automounter deciding to try unmounting an NFS filing system while the server is unresponsive. The result is that that all accesses to the mounted filing system stall indefinitely (but interruptibly) even after the server is back.
This is low-frequency for any particular server reboot, but still sufficiently common that we try to schedule things so that we can follow up the server reboot by a client reboot if it strikes.
We use amd to automount nfs on all of our platforms (hp, sun, sgi, aix linux) and always have problems when the filer becomes unavailable. So every time we have to perform maintenance on one of our filers, I have to reboot close to 100 systems. Does anyone know of a good nfs automount product? Do the automounters that come shipped with their respective OS work better than those purchased from outside software vendors? What other vendors offer legitimate nfs automount products?
Hey, what's that command to have ONTAP print out uptime stats? availsomething? I'm curious to see what our stats have been over the last three years.
uptime
-
Ambrose_Earle＠shamrockfoods.com -
Chris Lamb -
Chris Thompson -
Geoff Hardin -
Jeff Burton -
jeff.mery＠ni.com -
Stephane Bentebba