jay@hqda-ai.ARPA (Jay Hiser) (07/15/88)
I suspect that I could get a version of last that I could compile on my CCI 6/32. That would be useful, but all I really need is a list of each users LAST login -- not all of their logins. My goal is to identify obsolete accounts. I don't want to keep my wtmp info for 3 months at a time -- it collects about 50k a day. Given only my own resources, I'd write a shell script that runs who on wtmp every day, merges it with an existing list, and keeps only the latest entry for each user. That seems to be the quickest and easiest way to collect my needed info in a useful form. Not wanting to reinvent the wheel, though, I'm looking for a shell script that someone else has already written (probably a neater job than I'd do, too). So, if you've got something that your proud of and is useful at your sys, or some relevant info on this problem, I'd appreciate hearing from you. Thank you,
bowles@lll-crg.llnl.gov (Jeff Bowles) (07/15/88)
In article <7286@hqda-ai.ARPA> jay@hqda-ai.ARPA (Jay Hiser) writes: >I suspect that I could get a version of last that I could compile on >my CCI 6/32. That would be useful, but all I really need is a list >of each users LAST login -- not all of their logins. My goal is to >identify obsolete accounts. I don't want to keep my wtmp info for 3 >months at a time -- it collects about 50k a day. So turn on password aging and wait a couple of months! Then the people who don't use the machine will have "obviously old" passwords and you can work to delete those accounts. See passwd(4) for details. Jeff Bowles