APAR status
Closed as duplicate of another APAR.
Error description
The logic surrounding the error condition "AMgr: Unexpected error from MQPut: Message Queue is full", will cause Agent Manager to stop processing all agents currently in the queue. What should happen, is that agent manager should keep processing agents currently in the queues, so that when it catches up, it will recover and continue processing agents and new documents. It was relatively easy for this condition to occur due to spr TBOT3YRNYF. With that spr, once the queue was full and this condition was encountered, doing a Tell Amgr Sched, would show all sorts of agents in pending queues. The feeling is that agents currently in the queues should have been processed when in fact they didn't. Now that spr TBOT3YRNYF, is resolved, the condition is less likely to occur. However, with machines getting faster, and servers handling more db's, users, & agents, it is not unfeasible where this error condition will occur for real. To be prepared, the code should be modified to continue processing agents currently in queue, so that the message queue will be reduced, and more doc updates/mail deliveries will get added to the queue and continue processing. 99/07/23 08:37:08 AM AMgr: Unexpected error from MQPut: Message Queue is full. many many times........ 99/07/23 08:37:08 AM AMgr: Unexpected error from MQPut: Message Queue is full. > sh stat Agent.Daily.AccessDenials = 0 Agent.Daily.ScheduledRuns = 1418 Agent.Daily.TriggeredRuns = 211 Agent.Daily.UnsuccessfulRuns = 7 Agent.Daily.UsedRunTime = 2790 Seconds Agent.Hourly.AccessDenials = 0 Agent.Hourly.ScheduledRuns = 105 Agent.Hourly.TriggeredRuns = 5 Agent.Hourly.UnsuccessfulRuns = 1 Agent.Hourly.UsedRunTime = 108 Seconds > tel amgr sched E S 08:36 AM Today Agent-A | mta30 mail2/xxxx.nsf E S 08:36 AM Today Agent-A | mta30 mail2/xxxx.nsf E S 08:36 AM Today Agent-A | mta30 mail2/xxxxx.nsf E S 08:36 AM Today Agent-A | mta30 mail2/HYAA7792.nsf E S 08:36 AM Today Agent-A | mta30 mail2/HYAA4525.nsf E S 08:36 AM Today Agent-A | mta30 mail2/hkaa7726.nsf E S 08:36 AM Today Agent-A | mta30 mail2/IYAB2885.nsf E S 08:36 AM Today Agent-A | mta30 mail2/IYAB2917.nsf E S 08:36 AM Today Agent-A | mta30 mail2/IHAB3146.nsf E S 08:36 AM Today Agent-A | mta30 mail1/KSAB2904.nsf E S 08:36 AM Today Agent-A | mta30 mail2/gyab1520.nsf E S 08:36 AM Today Agent-A | mta30 mail2/IKAB2955.nsf E S 08:36 AM Today Agent-A | mta30 mail2/ITAA5855.nsf E S 08:36 AM Today Agent-A | mta30 mail2/IYAB2928.nsf E S 08:36 AM Today Agent-A | mta30 mail2/IYAB3167.nsf E S 08:36 AM Today Agent-A | mta30 mail2/kmaa8273.nsf E S 08:36 AM Today Agent-A | mta30 mail1/KJAB3104.nsf E S 08:36 AM Today Agent-A | mta30 mail1/KTAB3071.nsf E S 08:36 AM Today Agent-A | mta30 mail2/KAAB3162.nsf E S 08:36 AM Today Agent-A | mta30 mail2/KSAB2904.nsf E S 08:36 AM Today Agent-A | mta30 mail2/fjtest23.nsf E S 08:36 AM Today Agent-A | mta30 mail2/KYAB2920.nsf E S 08:36 AM Today Agent-A | mta30 mail1/KHAB2887.nsf E S 08:36 AM Today Agent-A | mta30 mail2/KIAB2860.nsf E S 08:36 AM Today Agent-A | mta30 mail1/MTAB3039.nsf V M Agent-ZZZ | fmall mail2/tmab0062.nsf V M Agent-ZZZ | fmall mail2/TSAB3171.nsf V M Agent-ZZZ | fmall mail1/AKAA9356.nsf V M Agent-ZZZ | fmall mail2/TOAB3142.nsf V M Agent-ZZZ | fmall mail1/WKAB2964.nsf V M Agent-ZZZ | fmall mail2/WKAB2964.nsf V M Agent-ZZZ | fmall mail2/yraa7449.nsf V M Agent-ZZZ | fmall mail1/YAAB2870.nsf V M Agent-ZZZ | fmall mail2/YAAB2870.nsf ents written: 3 I have not reproduced, but in theory the following should work: - Have a mixture mail trigger, doc update, and scheduled agents running on a server - Increase the values for AMgr_DocUpdateAgentMinInterval, AMgr_DocUpdateEventDelay, AMgr_NewMailAgentMinInterval, AMgr_NewMailAgentEventDelay to a value large enough to allow many doc updates. Perhaps 120 minutes - Create and send over 65k individual mail messages across 1 or more db's on the same server - Update over 65k individual docs across 1 or more db's on the same server - You should see the error "AMgr: Unexpected error from MQPut: Message Queue is full" - At this point do a "Tell Amgr Sched", and you will see agents stuck in queue without progressing
Local fix
Quit AMGR and restart or restart server
Problem summary
Problem conclusion
Temporary fix
Comments
This APAR is associated with SPR# SVRO4ATNLE. This APAR is a duplicate of LO34269
APAR Information
APAR number
LO41194
Reported component name
NOTES/DOMINO 7X
Reported component ID
5724E6200
Reported release
802
Status
CLOSED DUB
PE
NoPE
HIPER
NoHIPER
Special Attention
NoSpecatt
Submitted date
2009-06-05
Closed date
2010-12-21
Last modified date
2010-12-21
APAR is sysrouted FROM one or more of the following:
APAR is sysrouted TO one or more of the following:
Fix information
Applicable component levels
[{"Business Unit":{"code":"BU055","label":"Cognitive Applications"},"Product":{"code":"SSKTMJ","label":"Lotus Domino"},"Component":"","ARM Category":[],"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"8.0.2","Edition":"","Line of Business":{"code":"","label":""}}]
Document Information
Modified date:
21 December 2010