| Title: | VAX DBMS |
| Notice: | THIS NOTESFILE IS NOT A FORMAL SUPPORT CHANNEL |
| Moderator: | SCARY::CHARLAND |
| Created: | Thu Feb 20 1986 |
| Last Modified: | Tue Jun 03 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 2642 |
| Total number of notes: | 11044 |
We are running a manufacturing system under DBMS V6.0-2, and we've
run into a few strange problems recently that we can't seem to resolve.
Any ideas or suggestions on these ?
1) Encountered a bugcheck with
***** Exception at 0006212E : MBX$EXCHANGE + 0000026A
%DBM-F-MONITOR, database monitor is not running
after running DBO/CLOSE/ABORT=DELPRC on node A. Another node, B,
detected these deleted processes as "failed users" and started 19
recovery processes.
Later, the monitor terminated these recovery processes with bugcheck dumps.
During this time, all monitors were up and running on all nodes.
(Also wondering why node A didn't do the recovery...)
Could mailboxes be full causing failure of recoveries to communicate with
the monitor ?
2) Failure to attach to db with "exceeded max # of users"
DB user limit is 300 users.
At time of problem there were only about 100 users attached on
a particular node.
Are there any known problems in this area ? Or scenarios where access
to the database would be restricted ?
We also experienced a similar problem with Rdb.
3) AIJ file create fails with "users still attached"
To perform backup, DBO/CLOSE/ABORT=DELPRC and DBO/CLOSE/ABORT=FORCEX
are executed to kick users off to get a quiet point.
Then, code is executed to create a new AIJ file, but AIJ file create
fails with "user still attached". DBO/DUMP/USERS showed no users
attached...
This problem happended on three consecutive days. We restrict access to the
database through a common DB login procedure, and while it may be possible
that a user bypassed this mechanism, we thought we should verify whether
a known problem exists.
Thanks,
Dan McAuliffe
awecim::McAuliffe
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 2607.1 | Some questions... | BOUVS::OAKEY | I'll take Clueless for $500, Alex | Thu Mar 13 1997 10:39 | 39 |
>> <<< Note 2607.0 by ORAREP::AWECIM::MCAULIFFE >>> >> -< DBMS V6.0-2 problems/questions >- >>1) Encountered a bugcheck with >> >> ***** Exception at 0006212E : MBX$EXCHANGE + 0000026A >> %DBM-F-MONITOR, database monitor is not running >> >> after running DBO/CLOSE/ABORT=DELPRC on node A. Another node, B, >> detected these deleted processes as "failed users" and started 19 >> recovery processes. Who encountered the MBX$EXCHANGE messages? The db users on A or B? What were these users trying to do when they saw this message? >>2) Failure to attach to db with "exceeded max # of users" >> >> DB user limit is 300 users. >> At time of problem there were only about 100 users attached on >> a particular node. How many users total? You say 100 users on a particular node which implies to me that there might be other users on other nodes. Also, this is database users so if any process has more than 1 bind, each bind counts as a database user. >>3) AIJ file create fails with "users still attached" How are you creating the journal? What is the exact message? Single extensible or circular? >> To perform backup, DBO/CLOSE/ABORT=DELPRC and DBO/CLOSE/ABORT=FORCEX >> are executed to kick users off to get a quiet point. Database backup or journal backup? You shouldn't have to close the db to back up either. | |||||
| 2607.2 | It's been bugged | M5::LWILCOX | Chocolate in January!! | Wed Mar 19 1997 16:23 | 5 |
For whatever it might be worth, I have submitted two escalations on this problem, one for DBMS and one for Rdb. Both are assigned to the same engineer. Liz | |||||