WHAT TO DO WHEN SIRSI IS DOWN IF YOU ARE ***not* SIRSI SUPPORT STAFF

***For ease you may find it helpful to PRINT this WEBPAGE OUT***

1. If you can't find Linda Pearce or Nella Lall or David Brown, tell Mary Westell that the problem exists.
2. Ask the caller if they can connect to other services like email, the test server, etc. If they can NOT, then it's probably a network or hub problem -- NOTIFY DON KIRKHAM.
3. If they can connect to everything but seter, then:
4. See if you can connect or not
5. If you try telnet do you get a login prompt?
6. If you don't get a login prompt, the whole server may be down, so you should CALL OPERATIONS to see if they see any message at the server console.
7. OPERATIONS: ph. 220-5445. If the server has crashed, Operations will reboot it. Give your name and number so they can call you after it is back up again.
8. Call the numbers below to apprise them of the system status:
9. Barrie Main - 220-4424
10. Joseph Yip - 220-6218 (Email: yip@ucalgary.ca)

DESKS

1. Circulation desk: 220-5224/5963
2. Lucille Wallace: 220-3758
3. LRS: 220-5966 (Remind them they can access neter.lib.ucalgary.ca for the catalogue.
4. Doucette: 220-3982
5. David Brown: 220-6295
6. Susan Jenkins in Acq: 220-5958
7. Gloria Dingwall in Ser: 220-3325
8. Wanda Rottenfusser in Bib.Services: 220-6161
9. Carol Baker in Bib.Services: 220-6165
10. Medical library circulation: 220-6855
11. Law circulation: 220-3727
12. Gallagher library: 220-6042/6043
13. Haskayne School of Business: 220-6113/7720
14. Music circulation: 220-3610
15. Cindy Murrell: 220-6162

After a reboot, go into seter immediately and initialize all the users and all the unicorn stations. If you don't know how to do this, call the Sirsi helpdesk at 1-256-704-7001 and ask them to do it.

If the reboot does not succeed in bringing the system back up, call the Sirsi helpdesk

FOR SIRSI SUPPORT STAFF ONLY:

• If the whole system has ground to a halt, look for runaway processes. If found, kill them.
• If there are no runaway processes and you can't find anything else wrong, do a halt-and-run.
• Reinitialize the servers, stations and users before re-running the servers.
• If only Webcat is down, the http daemon is probably down. Check and restart it.
• After the system is back up, check to see if any reports were running when the system crashed, because they may have to be re-run.
• Then check to make sure none of the following programs that run via the crontab were running during the crash, because these are the ones that call the API server, meaning that Unicorn had to be up during their processing:
  idbatch_cron: the weekly ID/User update
  emailupdt_cron: the nightly student user email address update
  commit.sh: Weekly monograph order commitments and cancellations interface with FRS
  payments.sh: Weekly monograph order payments interface with FRS
  verify_cron: Checks whether the lib. user's verification process is running and, if not, restarts the program (also on neter)IT DOESN'T MATTER IF THIS ONE WAS RUNNING OR NOT WHEN THE SYSTEM CRASHED