[prev in list] [next in list] [prev in thread] [next in thread] 

List:       opensolaris-lvm-discuss
Subject:    Re: [lvm-discuss] Slow metadb and metastat on SXCR Build 55
From:       Pete Bentley <pete () sorted ! org>
Date:       2007-02-02 17:00:04
Message-ID: 45C36E14.1050702 () sorted ! org
[Download RAW message or body]

Eric Ham wrote:
> So I tried running metadb by itself in another shell and after 10 minutes I hit \
> Ctrl-C since the output never finished. I also tried running metastat and it \
> finally finished after 31 minutes: [...]
> Finally, I did a truss on metadb and have attached the output file "truss.txt" \
> after running it for 60 seconds and using Ctrl-C to quit.  The little bit I can \
> comprehend of this output seems that perhaps there is a loop of sleeping and then \
> trying to open /dev/udp: 
> nanosleep(0xFFBFF0C8, 0xFFBFF0C0) (sleeping...)
> nanosleep(0xFFBFF0C8, 0xFFBFF0C0)               = 0
> open("/dev/udp", O_RDONLY)                      = 5
> 
> The system in general seems responsive to other commands, just metadb and metastat \
> are super slow. Any thoughts on what might be causing this? Or other commands I can \
> try to use for debugging this?

Is it possible that rpc.metad isn't running, perhaps because 
svc:/network/rpc/meta:default is offline?  The truss might then be 
showing that metadb is sitting in a loop trying to contact it via 
RPC/UDP and timing out...

Pete.


_______________________________________________
lvm-discuss mailing list
lvm-discuss@opensolaris.org


[prev in list] [next in list] [prev in thread] [next in thread]