| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 1051.1 | | MUZICK::RITCHIE | Elaine Kokernak Ritchie, 227-3089 | Mon Apr 01 1996 17:18 | 1 |
| Which version of Scheduler is being run?
|
| 1051.2 | | MUZICK::RITCHIE | Elaine Kokernak Ritchie, 227-3089 | Mon Apr 01 1996 18:25 | 11 |
| The node name change doesn't affect the agent starting up. The only
thing that will keep an agent from starting is incorrect network setup.
Your customer needs to be sure that DCE and the Bind server are setup
correctly.
In reference to changing a node name, once the Scheduler is up and
running, you need to modify any jobs which have been told to run on
the old node name, so they will run on the new node name. This should
be the only change that needs to be done.
Elaine
|
| 1051.3 | | MRBASS::PUISHYS | Project Leader Scheduler V3.0 for Digital UNIX | Thu Apr 04 1996 17:34 | 21 |
| Hi Glenn,
Is it the V3.0 unix agent that will not start? Does the log file show
a core dump?
If so do the following:
on the agent:
ls -l *.ckp
ls -l *.unsent
if you find any of these files that are zero (0) blocks delete them.
restart the agent. If it starts ok on all the servers that use that
agent you must do a verify server, to get the jobs back insink
with the agent.
if you deleted any files you lost the last run info on a job, or
currently run info of any jobs.
The verify server will fix the jobs, but htey rerun.
bob
|