screen-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Screen crash - not really :^)


From: Andy Goth
Subject: Screen crash - not really :^)
Date: Wed, 20 Apr 2005 00:01:16 -0400
User-agent: Mutt/1.2.5.1i

Screen crashed a little while ago.

I was running mutt inside screen through ssh, and it halted
(coincidentally, while "Wuff  ----  Wuff!!" was on the display).  I
ssh'ed in again and surveyed the situation:

address@hidden|~]$ ps -Uandy -x | grep -i screen | grep -v grep
 7099 ?        S      3:28 SCREEN
 9521 pts/26   S      0:00 screen -x
13474 pts/19   S      0:00 screen -x

0% CPU time, according to top.  Hmm.  Blocked?

And then it uncrashed. :^)

While doing other work, it came back to life and I failed to notice the
exact moment.  In fact, I'm using the very same screen session to write
this email.

But during the time it was, I dunno, whatever it was, I joined this list
to gripe about the fact that this (the crash, not the uncrash, heh) has
happened before several times.  True, it's quite rare (once every couple
months of constant use), but every time it occurs, I lose data.  (Thank
God for vim swapfiles!)

The crash (let's be more specific: the SIGKILL-requiring freeze) has
only happened so far on this one host, ducks.  So I asked the sysadmin
to upgrade.  Eventually he did, and eventually the newer version met the
same fate.  Assuming that the whole thing was just some Red Hat
(per)version, I installed my own copy in ~/package/ (a la sp-foreign),
which I am using now.  So far, the current incident is the only oddity I
have experienced.  (I think.)

So what I'm *really* asking is, are there any known problems with the
Red Hat packages of GNU screen?

About the crash/uncrash, it might just be lost packets (which happen
relatively frequently, as my computer has a semi-broken serial port).

-- 
Andy Goth  +  address@hidden  +  http://ioioio.net/




reply via email to

[Prev in Thread] Current Thread [Next in Thread]