[Veritas-bu] bpstart behaviour
2005-12-19 08:58:19
I am running netbackup 5.0mp3 on Solaris 9. Attached to a L700 with 5
LTo2 tape drives fiber attached.
ok,
I have an issue where one of my media servers cant see the 3 shared
drives all of a sudden. sgscan comes back with none and cfgadm -al shows
them as being unavailable. Why i dont know and am trying to work this out.
As a stop gab i added this media manager to a standard policy on the
master server which backs up clients over the network so it would back
this media server up over the network so at least we are getting backups.
As this media server is also an oracle DB server i need it to not start
backing up until past 12:30am and the normal policy starts around
10:30pm. So i created a bpstart_notify.Standard on the media manger
server in the correct place and put sleep 10800 in it so it would sleep
for a few hours. I then started to get 74 messages of bpstart timeout.
To fix this i put bpstart_timeout = 11000 in the bp.conf on the media
manger and then had to put it on the master server. This annoys me
slightly as this will now affect all bpstart timeouts for each client.
Is there not a client bpstart timeout option because i couldnt find one
in the admin guides.
Anyway this seems to have stopped the timeout status but i am now
getting a status 41 which is network timeout. Has anyone any ideas? I
know the network connections are all fine as running a manual backup or
even a scheduled one without the sleep statement works fine.
Why the drives have stopped being visible i have no idea as no changes
have happened on the system.
Thanks
|
<Prev in Thread] |
Current Thread |
[Next in Thread>
|
- [Veritas-bu] bpstart behaviour,
Dave Markham <=
|
|
|