Hi Peter.
I have never found this to be much of a problem.
Until you get that message in dsmsched.log that says "SENDING RESULTS TO
SERVER", the server knows the client has not finished backing up that
day.
In my Win95 and WinNT client options files I have:
MAXCMDRETRIES 99
RETRYPERIOD 20
QUERYSCHEDPERIOD 24
If the server goes down during a backup, the client gets a TCP/IP
connection failure. Because RETRYPERIOD is 20, the client retries the
connection every 20 minutes. When the server comes back up, the client
reconnects and starts sending again.
Same thing on the client end - if the client goes down, when it comes
back up again its Scheduler reconnects and starts sending, because it
never finished the backup the first time.
But I don't think you will have much luck as long as you have your event
windows limited to 1 hour. That doesn't give you much time to reboot
either a failed server or a failed client, and get your backup
completed. I think if you will expand your window, the clients will
pretty much reschedule themselves.
A good way to test this is just to start a backup on one of your
clients, and pull the plug on your network connection. When you reboot,
look at the server activity log, and the client dsmsched.log, and you
can see what they are trying to do.
=======================================================================
Wanda Prather
Johns Hopkins Applied Physics Lab
301-953-6000 X8769
wanda_prather AT jhuapl DOT edu
"Intelligence has much less practical application than you'd think."
- Scott Adams/Dilbert
=======================================================================
> ----------
> From: Peter Thomas[SMTP:Peter_Thomas AT manulife DOT com]
> Sent: Thursday, June 05, 1997 2:16 PM
> To: ADSM-L AT VM.MARIST DOT EDU
> Subject: ADSM / Scheduled Backups
>
> From: Peter Thomas on 06/05/97 02:16 PM
>
> If I have a client machine that has a backup scheduled to happen at
> say
> 11:00 PM, the backup commences but for some reason the server
> "crashes"
> before the backup completes, is there any way of having ADSM
> "reschedule"
> that backup so it runs again upon completion?
>
> Right now, I have the clients operating in SCHEDMODE POLLING, with
> TCP/IP
> as the communications protocol, a Query Schedule period of 1 hour. The
> Scheduled events are defined to have a 1 hour duration and occur
> daily.
>
> What can I do to ensure that if the client crashes during the
> execution of
> the event and restarts itself (along with the scheduler) that the
> event is
> retriggered?
>
> Thanks
>
> Peter Thomas
> Manulife Financial
> Phone: (416)963-7371
> Fax: (416)926-5454
>
|
