High CPU Usage for TCPIP$HR_MIB Process

[D.W. Johnson]

We are just now completing the configuration of a new Integrity RX6600 on which we are running OpenVMS V8.3 and HP TCP/IP V5.6.

This is a new turnkey application server. The application vendor says they don't normally set up the TCPIP$HR_MIB process on their servers. (I believe this is one of the SNMP processes.) Yet, this process cranked up in the last 24 hours and is consuming a lot of CPU. We rebooted the server to correct the situation. The process has not reappeared. What causes this process to fire up?

Thanks,
D. W. Johnson

[Hoff]

I haven't seen a particular resolution of this, though there have been a few previous reports of this process (the host resources snmp mib) looping.

And there are various fixes for loops listed in various older ECOs.

This at least as far back as V5.3 or so, and possibly further back.

The IP manuals make a few tangential references to errors, and point to the use of SYS$STARTUP:TCPIP$SNMP_SHUTDOWN.COM and of TCPIP$SNMP_STARTUP.COM

If you are current on ECOs for V5.6 (and I haven't looked to see if there is one out) and this loop happens again, I'd tend to use show process /continuous or such, and collect a series of PC values for the looping. If it'll accept this command, I'd also try aiming a set process /dump=now /id=pid command at the looping process.

Then ring up HP.

[Jan van den Ende]

D.W.

To start,

WELCOME to the VMS forum!

As for your issue, this may or might not apply, but there IS an recent new networks patch available.
I have not yet gone over the release notes, but it might be interesting. Look them over, if this somehow seems to remotely chime, _DO_ give it a try (and report back, please)

fwiw,

Proost.

Have one on me.

jpe

[Wim Van den Wyngaert]

The process has a log file tcpip$hr_mib.log.
There might be something in it ...

Wim