> On Oct 19, 2017, at 1:39 AM, Vernooij, Kees (ITOPT1) - KLM > <[email protected]> wrote: > > We have all syslogs scanned every morning and interesting messages extracted > and emailed. > > Kees. Kees,
That is always an issue to me as (at least I) don’t know exactly what I am looking for. For example: I happen to know the silo has tape addresses between 8900 and 8963 So I have to look at any messages about those devices or some message that the SILO software sent to the console around the time in question. There has to be a narrowing issue and that is really all I need. Consoles are flooded with messages and if take the route of between 0100 and 0115 you spend to much time trying to diagnose items that are not relevent. The same thing with system dumps, you really need to use the fuzzy factor of there was an S0C4 in module ifg0999a (example) I always look say 5 minutes before and maybe 2-3 minutes after. If an automation package just automatically sends me 2000 lines before and after it takes more time to scroll 4000 lines. I have seen a case where the operator cancelled a stc and it took 4 hours for the stc to acknowledge the cancel and then 2 hours for the sec to end. Please forgive the typo’s. I have yet to see any automation tool that can give me the information that I think is germane to the issue. If you know of one please let me know at it would save me a fair amount of time on problems. Also the automation package has to know to run IFcerep1 for the time surround the issue *AND* be able to see and SNA type errors (From the database that its kept). I don’t think it is possible. But I am willing to learn. Ed ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
