inactivity causing abandon by nsrexec

[dreagarcia]

Overview:
Learning backup for about 4 - 5 months now. Everything was working fine until our development servers and backup server were migrated to a new domain, renamed and re-IP'd. Now every night some (but not all) backups fail with a variety of errors such as:
<servername1>:\has been inactive for 30 minutes since <datetime>
<servername1>:\is being abandoned by nsrexec

<servername2>:C:\1 retry attempted
<servername2>:C:\No full backups of this save set were found, performing full backup
<servername2>:C:\save: RPC error: Unable to send

<servername3>:All 1 retry attempted
<servername3>:All Host <backupserver> cannot request command execution
<servername3>:All <datetime> nsrexec:Host <backupserver> cannot request command execution
<servername3>:All Permission denied

That is about all the different messages. I have already done the following:
1. Checked port speed settings. These were all OK except for 2 of the servers, and there are about 7 that fail every day.
2. Verified backup server and client host files.
3. Verified aliases in networker client configuration.
4. Stopped & started Networker Remote Executive service.
5. Pinged to verify network connectivity, all successful.
6. Client shortcuts have -s <backupservername> configured.
7. res file has the correct backup server name.

Oh yea, the backup server is running 2000 server, and the clients are running either 2000 server or NT 4.0 sp6a. Networker version 6.0.1.
What am I missing? This has been going on for over a week, and it's giving me the blues! I have been copying files manually to other servers to provide a little security to my development team, but I need to get my backups working again!

 

[dreagarcia]

Oh yea I almost forgot. The event viewer log on the backup server has an error message with a date of 8/2/02, saying something about a device has a bad block. There is also an error,&quot;Networker media; (alert) Volumes in jukebox <name> have passed the low water mark.&quot;
What does that mean?
Any help will be appreciated!

 

[ulsj]

Hi!

The low water mark is a way for networker to tell you that the free space on tapes in the jbx is low. I think the default setting is 7GB ant it can be change via the jbx properties.

/Ulf

 

[soder]

Hi,
If you recive something like this..... then you should upgrade your server to 6.0.3 or 6.1.1 since this is a known bug. (LGTpa30132)

Error encountered by NSR server `NW1': Bad or missing
record: saveset ID 1144599809, low water mark
139303368, current offset 139283040.

About that &quot;unable to send&quot; messages you recive, try to raise the inactivity timeout value (group value) to 90 or something like that.

/D

 

[dreagarcia]

Thanks ulsj and soder,
I tried raising the inactivity timeout and I noticed it was at 30 minutes, so I raised it to 60, but I still got the message that it tried for 35 minutes and then abandoned. That makes me think there is some other time setting I am not aware of.
I have asked for an upgrade from the production people, they are pretty good about responding to requests like that. I'll let you know if that works.
The low water mark raises another question, which is not answered by the thread posted today about re-using tapes. Can I keep re-using the same 2 sets of tapes over and over, as long as I recycle them? I will be getting a 3rd set tomorrow so I can keep a longer history.