[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Condor-users] Restarting comms with XP nodes (SP2 UDP bugette)
- Date: Thu, 23 Jun 2005 09:51:05 -0500
- From: Zachary Miller <zmiller@xxxxxxxxxxx>
- Subject: Re: [Condor-users] Restarting comms with XP nodes (SP2 UDP bugette)
On Thu, Jun 23, 2005 at 10:05:11AM -0400, Bryan S. Maher wrote:
> > for this, you need to also define COLLECTOR_SOCKET_CACHE_SIZE.
> Yeah, did that too. Sorry I wasn't more clear. I followed the
> procedure in the manual. I put this in my config files:
> COLLECTOR_SOCKET_CACHE_SIZE = 99 and
> UPDATE_COLLECTOR_WITH_TCP = TRUE
> I pushed the updated config files out to my master/collector and to two
> test execution/submit nodes. I issued condor_reconfig to all three. I
> immediately began receiving the "...via TCP; ignored" errors. I tried
> restarting condor on the master first, then on the two execution/submit
> nodes. The error persisted.
strange. i am unable to replicate this. are you sure the reconfig/restart
was successful (and didn't get PERMISSION DENIED or anything)? also, you can
find out for sure if the collector recognized the setting by turning on
D_FULLDEBUG for COLLECTOR_DEBUG. you should see a line like this in the
6/23 09:32:23 (fd:8) Using a SocketCache for TCP updates (size: 99)
otherwise, you'll see a line like this:
6/23 09:31:25 (fd:8) No SocketCache, will refuse TCP updates
if you want to try that, it will help me figure out what's going on.
> Thanks for the "session" info. I keep my entire pool at the same
> revision level so there is no mixed versioning going on here. Everyone
> is running 6.6.9 currently. I'm curious, you said that if the
> collector is restarted, it would invalidate any open sessions. Does the
> same hold true in reverse? That is, if an execution node is restarted,
> is there a potential for the same problem?
not in the case of the Collector (it doesn't contact the startds) but the
Negotiator would try to contact a startd if a job was run, and then would fail
the first time because of the missing session. then the startd informs the
negotiator of the invalid session, and next time (next negotiation cycle) the
negotiator will start a new session and everything will be back to normal.
> I'll try setting SEC_DEFAULT_NEGOTIATION = OPTIONAL and see if that
it should, so long as you aren't using any strong authentication, crypto,
or integrity checks. they all require the use of sessions.