[Search for users]
[Overall Top Noters]
[List of all Conferences]
[Download this site]
| Title: | DECmcc user notes file. Does not replace IPMT. |
| Notice: | Use IPMT for problems. Newsletter location in note 6187 |
| Moderator: | TAEC::BEROUD |
|
| Created: | Mon Aug 21 1989 |
| Last Modified: | Wed Jun 04 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 6497 |
| Total number of notes: | 27359 |
5683.0. "MCC event sink problem" by TROOA::GREENALL () Tue Oct 26 1993 15:24
Well folks, Yet another MCC problem.
This time, its a problem with the MCC collector event sink. It
will suddenly stop collecting events. The process still exists, but
the decnet object disappears from the system. This then causes breakin
events every time the event sink tries to send an event. I did qar
this (before the cut off date), and the qar was promptly ignored.
This is an urgent issue, as it is causing events to be lost at a
customer site, which we at digital have outsourced and are managing
thru DECmcc.
the only clue I have are messages like this in the mcc_evc_sink.log
file. I will be raising this problem to a CLD.
26-OCT-1993 15:20:02.00, Event #3 sent to MCC event manager
26-OCT-1993 15:20:02.60, Connected to COLLECTOR
>>Error receiving COLLECTOR event,status = 8420, 26-OCT-1993 15:20:04.60
26-OCT-1993 15:20:04.61, Disconnected link to COLLECTOR
26-OCT-1993 15:20:04.61, Disconnected link to COLLECTOR event sink
26-OCT-1993 15:20:04.61, Collector DECNET sink thread 65542 terminated
Alerting management thread, no more active sink thread>> Error waiting for internal event, at step 5, status = 52877226, 26-OCT-1993 15:20:08.27
26-OCT-1993 15:20:08.31, Terminating sink management thread
26-OCT-1993 15:20:08.31, Terminating COLLECTOR sink management thread,status = 52877226
26-OCT-1993 15:20:08.32, Terminating COLLECTOR event sink
%MCC-E-ALERT_TERMREQ, thread termination requested
SYSTEM job terminated at 26-OCT-1993 15:20:09.68
Accounting information:
Buffered I/O count: 442 Peak working set size: 6411
Direct I/O count: 211 Peak page file size: 15215
Page faults: 3027 Mounted volumes: 0
Charged CPU time: 0 00:00:28.70 Elapsed time: 0 03:56:25.00
| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 5683.1 | Does it make you feel better to know.......... | PLUNDR::LOWEG | WANTED!! A modern day Robin Hood. | Fri Oct 29 1993 07:33 | 8 |
|
How do you workaround this. It might (might) be the same strange
problem I am experiencing. The fix seems to be to re enable the
collection_am.. see note 5487 if you wish to feel a lot better knowing
that support people are struggling with the same process.
Gary Lowe UK MCS Nets&Comms ( Ultrix MCC (PNM200))
|
| 5683.2 | versions | TROOA::GREENALL | | Fri Oct 29 1993 11:06 | 6 |
|
Thanks. I realized though that I should have put some versions on my
note in .0 It is mcc V1.3 running on VMS 5.5-2
rich
|
| 5683.3 | | PLUNDR::LOWEG | WANTED!! A modern day Robin Hood. | Fri Oct 29 1993 11:27 | 6 |
|
Do you have a workaround Rich ???
Gary.
|
| 5683.4 | somewhat | TROOA::GREENALL | | Mon Nov 08 1993 16:11 | 13 |
|
sorry about taking so long to get back to the note, but
Nope, no workaround. The only thing that I do now is have a DCL
procedure check for the existance of the object, and blow away the
intrusions and restart the object. This procedure checks every 30
seconds.
The only problem, is that this is like closing the barn door after the
horses have escaped. Alarms will be lost, but at least this somewhat
limits them.
|