Hello managers, Here is the original question: While running uptime on our e4500 Sun I get the following printout: # uptime 11:06am 8 users, load average: 2.48, 2.90, 2.70 While on our other servers which also have been upgraded to Solaris 8 w= e get: # uptime 11:07am up 4 day(s), 23:53, 1 user, load average: 0.00, 0.00, 0.01 Does anyone know why the amount of days (uptime) does not show? The answer was that I recently mv and compressed the utmpx/wtmpx files which uptime depends on. I want to say thanks for the answers I recieve= d, Below is a summary of responses: Kevin Buterbaugh ? When's the last time your /var/adm/wtmpx file was cleaned out (keep in mind, that may be done automatically by a cron job)? Uptime determines= how long the system has been up by reading the last "reboot" entry in the w= tmpx file, so if it's been truncated, the uptime command cannot determine ho= w long the system has been up. We try to make it a habit to manually truncate the wmptx file prior to rebooting our servers so that we keep = the wtmpx file from getting too big while maintaining the uptime informatio= n. HTH... Joel Whitmer ? Sounds like you recycled your utmpx/wtmpx files recently? Ronald Loftin ? Either the box rebooted when you weren't looking, or you may have been hacked. Eric Shafto ? Do you have a log rotator that truncates /var/adm/wtmp? Koos van den Hout ? It uses the locale to find out how to print the time and days of uptime= . Maybe something is wrong with the locale settings/files. Grant Schoep ? Not sure, but I am guessing the one machine is getting a different upti= me. Do "which uptime" to make sure your not getting some built in shell upt= ime, or some ucb uptime, or something like that. Chris Josephes ? If I'm not mistaken, the uptime command gets the uptime statistic from = the utmpx/wtmpx files. Mark Hargrave- Try the full path name: /usr/bin/uptime You could be hitting another version of "uptime". Greg Ulyatt- Is uptime the same? Have you tried the /usr/bin/w command? It should also give the uptime. Last reboot should also give you the last reboot. If the proble= m persists, you may want to look into corruption in the /proc filesystem, that's where most of this info is kept... other than that, the only thing I can suggest is that (if) it's a MP system, perhaps there have been cpu's reset during the uptime? =Received on Thu Aug 2 14:36:57 2001
This archive was generated by hypermail 2.1.8 : Wed Mar 23 2016 - 16:25:00 EDT