[96276] in RedHat Linux List
Re: top is not working
daemon@ATHENA.MIT.EDU (Nikki Cook)
Mon Oct 26 06:27:16 1998
Date: Mon, 26 Oct 1998 06:26:58 -0500
To: redhat-list@redhat.com
From: Nikki Cook <sunny@mail.suntrix.com>
In-Reply-To: <005001be0042$f40eba40$0200a8c0@eric>
Resent-From: redhat-list@redhat.com
Reply-To: redhat-list@redhat.com
Hi Eric,
Thanks for the reply. I was beginning to think my post had gone un-noticed.
First, the network is fine. I can admin all or reach all machines on our
network without problems, nothing different or new, it's business as usual.
I can put monitors on each ( we normally run those RH boxes monitorless )
and login locally fine. Users can login in fine. Users didn't report any
problems, nor did I see any in /var/log/messages. I've even logged in
using rsh, no problem. Login works (when enabled), ssh works, both pre and
post incident. =20
Box#2 does show the same "bad data /var/run/utmp". I haven't done a
shutdown with it, it's a vhost webserver and it's functioning just fine,
except for top not running. Uptime is 66 days. I decided I wouldn't try
and fix it till I found an answer.=20
Box#3, which I haven't mentioned before, didn't exhibit the same problem.
Lemme give a clearer picture of the machines I was logged into at the time
this happened:
My admin machine =3D Win95 running Exceed using KDE=20
ssh -> Shell machine =3D RH 5.1 (2.0.34)
ssh->Terminal Server =3D Linux RH 4.2 (2.0.34) <- not effected
ssh->Web Server =3D RH 5.1 (2.0.34)
I think that when this occurred and I hit the power button on the shell
machine, it of course rudely dumped my Exceed Xserver session on each
machine. So the utmp data became confused as to who was logged in/out.
The common factor is RH 5.1 since both those machines have the same
symptoms and the RH 4.2 box doesn't.
RH 5.1 isn't cleaning up the data correctly, at least that's what it feels
like to me, but I still don't know if I'm on the right track, and even if I
were, what to do with the information to fix this. What I've already tried
hasn't fixed it.
If I knew what caused the original manifestation of the "racing" tail -f
screen, I'd have my answer, I'm sure. But nothing significant was in the
/var/log/messages. And last, if I didn't need top as an effective admin
tool to keep track of realtime usage, I wouldn't be asking for help on
this. Bottom line is, I do need it. Keeping track of users without it is
much harder, believe me.
Nikki
At 01:11 PM 10/25/98 , you wrote:
>Well it seems like a hardware or network problem. Perhaps the problem is
>with the machine you use to ssh. Were there any symptoms of problems from
>users of this machine previous to yours? Are you sure? Does the "other"
>5.1 box (box #2) exhibit the same problems? Can you log into the machines
>locally after "cleaning" them and do they exhibit the same problems?
>
>- Eric
>
>-----Original Message-----
>From: Nikki Cook <sunny@mail.suntrix.com>
>To: redhat-list@redhat.com <redhat-list@redhat.com>
>Date: Saturday, October 24, 1998 1:02 PM
>Subject: top is not working
>
>
>Ok, now I'm comfortable enough with how things are done here to present you
>all with my problem in hopes that you can help me.
>
>This is on a 5.1 box (orig 2.0.34 kernel from the CD) that has been running
>flawlessly since July until about a week ago. I normally admin that
>machine via ssh. I also normally tail -f messages because this box is a
>shell user box.
>
>The first thing I saw was the tail screen go nuts, like a panic or race. I
>couldn't do anything with the box to get control, nothing responded, so I
>hit the power button.
>
>It rebooted just fine, forced the fsck without any glaring problems. All
>daemons loaded, all croned user processes started, and everything looked
>just ducky 'till I ran top to check on their procesess. I noticed that
>Top's CPU % was about half what it normally took. I checked file size/date
>against another 5.1 box on our network, no difference.
>
>The next time I tried to run top, it threw a "bad data in /var/run/utmp" at
>me, and made that xterm unusable.
>
>After reading and digging for info, I rm'd /var/run/utmp and /var/log/wtmp,
>touched those files, logged out and back in. Top still wouldn't run. Now
>this is really weird, Top wouldn't run on that other 5.1 box either. Yes,
>I was logged into it as well via ssh when all this started.
>
>Back to box #1... I rm'd and touched the files again, rebooted, changed to
>single user mode and ran fsck on each partition. It fixed stuff, then I
>ran it a second time on each partition and all partitions showed clean.
>
>Logged in. Top ran 3 times, then started throwing the same "bad data"
>complaint at me.
>
>I checked /dev/ttyp*, didn't see anything weird there.
>
>If I've left lots out that you need to know to help me or point me in a
>direction, my apologies. Yep, I'm one of those "newbies" that needs
>somebody to take me by my sweaty li'l palm and explicitly and succinctly
>tell me how to fix this. I just don't have the know-how to sort this out
>and fix it on my own.
>
>
>TIA,
>
>Nikki
>
>
>~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> SunTrix Com Internet Services
>Daytona Beach, Florida
> PPP and Shell Accounts (904) 258-5434
>WEB Design webdesign@mail.suntrix.com
>http://www.suntrix.com
>WEBBnet IRC Network
>irc.webbnet.org | irc.us.webbnet.org
> ftp://ftp.suntrix.com | mail.suntrix.com
>
>~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
>
>--
> PLEASE read the Red Hat FAQ, Tips, Errata and the MAILING LIST ARCHIVES!
>http://www.redhat.com http://archive.redhat.com
> To unsubscribe: mail redhat-list-request@redhat.com with
> "unsubscribe" as the Subject.
>
>
>
>
>--=20
> PLEASE read the Red Hat FAQ, Tips, Errata and the MAILING LIST ARCHIVES!
> http://www.redhat.com http://archive.redhat.com
> To unsubscribe: mail redhat-list-request@redhat.com with=20
> "unsubscribe" as the Subject.
>=20
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
SunTrix Com Internet Services
=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 Daytona Beach,=
Florida
=A0=A0=A0=A0 PPP and Shell Accounts (904) 258-5434
=A0=A0=A0 WEB Design webdesign@mail.suntrix.com
=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 http://www.suntrix.com
=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 WEBBnet IRC Network
=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 irc.webbnet.org | irc.us.webbnet.org
=A0=A0=A0=A0=A0=A0=A0ftp://ftp.suntrix.com | mail.suntrix.com
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~=20
--
PLEASE read the Red Hat FAQ, Tips, Errata and the MAILING LIST ARCHIVES!
http://www.redhat.com http://archive.redhat.com
To unsubscribe: mail redhat-list-request@redhat.com with
"unsubscribe" as the Subject.
