SUMMARY: Sun systems hang about once a day.

From: Mark Costlow (edsr!tantalum!cheeks@uunet.uu.net)
Date: Wed May 15 1991 - 10:57:13 CDT


I received several responses in a very short time (as many have said, this
list is great). Here's a summary:

Thanks to:
        ajc@mcm.com
        finnell@community-chest.mitre.org
        uunet!oddjob.uchicago.edu!oconnor!gorpong
        uunet!sunne.East.Sun.COM!stern
        uunet!mcnc.org!ge-dab!ORCAD2.dnet.ge.com!THIER

"ajc@mcm.com", "uunet!sunne.East.Sun.COM!stern", and
"uunet!mcnc.org!ge-dab!ORCAD2.dnet.ge.com!THIER" all suggested Patch-ID#
100159-01 which deals with a problem with Sun's tcp-loopback mechanism --
it apparently has a tendency to hang the system if used a lot. I've
applied this patch, and am waiting for the system to crash or not.

finnell@community-chest.mitre.org suggested I look for cron jobs deleting
things from /tmp or elsewhere on a daily basis. I do have a couple of
these, but don't see the connection. I forgot to mention in my original
note that we are not using tmpfs (mostly because of all the grief we've
seen on this list).

uunet!oddjob.uchicago.edu!oconnor!gorpong pointed out that sendmail isn't
actually started by inetd. He's correct ... I had just pulled a daemon
name off the top of my head when I posted.

Thanks again for all responses!

>>Chx

--
cheeks@edsr.eds.com    or     ...uunet!edsr!cheeks



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:13 CDT