CHAMBERS_K@CHO006.CHO.GE.COM (Keith - Systems & Tools - 8*277-6132) (02/05/90)
We run a link to a remote site that we connect to three times a day using
a DECnet synchronous dialup link. We are getting MAIL through, but we are
seeing the following in the log files when the link is down - should it
die with a stack dump?
OpFile: DN_DE01_MASTER.LOG
Op
$ verify = f$verify(0)
Batch job 970, queue AT$BATCH on cpu CHO006.
Total time 12 secs (6 secs to LOGIN.COM + 6 for LOGIN.COM)
%PAS-F-HALT, HALT procedure called
%TRACE-F-TRACEBACK, symbolic stack dump follows
module name routine name line rel PC abs PC
00044E4F 00044E4F
DN_MASTER DI_FATAL_ERROR 1353 000003C0 0000599C
DN_MASTER DI_INIT 1366 0000015F 00006E73
DN_MASTER DN_MASTER 2806 00000096 00007072
SYSTEM job terminated at 2-FEB-1990 14:02:29.73
Accounting information:
Buffered I/O count: 182 Peak working set size: 913
Direct I/O count: 102 Peak page file size: 3114
Page faults: 946 Mounted volumes: 0
Charged CPU time: 0 00:00:09.22 Elapsed time: 0 00:00:19.50NED@HMCVAX.CLAREMONT.EDU (Ned Freed, Postmaster) (02/05/90)
The stack dump is normal is the link cannot be made. I don't suppress
stack dumps in PMDF channel programs as a rule -- they are extremely
informative when something is actually wrong.
You probably do not want to have delivery attempts made when the link
is down. If you mark the channel as slave only, no attempts will be made
to deliver messages queued to the channel. Then you can use MASTER.COM
to service the channel in a periodic job of your own that only runs when
the link is actually up. (See the comments at the end of MASTER.COM for
a description of the parameters.)
Of course, what you have now does work, so you may want to leave it alone.
Ned