LWN.net Logo

Linux vs Solaris stability

Linux vs Solaris stability

Posted Sep 20, 2003 0:11 UTC (Sat) by raph (guest, #326)
In reply to: Sun's Schwartz Speaks Out on Linux, SCO (eWeek) by captrb
Parent article: Sun's Schwartz Speaks Out on Linux, SCO (eWeek)

This sounds like the kind of thing that can and should be measured quantitatively. It wouldn't be cheap to do, but should be relatively straightforward. Just put both OS's on identical hardware, put an insane application load on it (near out-of-memory conditions, saturated network, etc.), and see how many crashes you get.

In the meantime, readers may find this survey from the sunmanagers mailing list interesting.


(Log in to post comments)

Linux vs Solaris stability

Posted Sep 20, 2003 1:31 UTC (Sat) by jmorris42 (subscriber, #2203) [Link]

Wow. And that on a list likely to be inhabited by Sun Worshipers. Very fair and balanced look at the world as it existed two years ago. And most of the problems identified have been addressed.

Linux vs Solaris stability

Posted Sep 22, 2003 8:19 UTC (Mon) by error27 (subscriber, #8346) [Link]

put an insane application load on it (near out-of-memory conditions, saturated network, etc.

At work I test x86 hardware with Linux. I have seen some load related bugs. We prefer the e100 driver to the epro100 for this reason. I've seen one or two SCSI driver problems that were related to load. But really load related bugs are rare. When you do find a software bug, you just fix it... For example last week, we hired a consultant to fix a kernel bug in handling large RAID arrays.

Basically, I don't think it is possible to compare the two operating systems that way. If you could trigger bugs that easily the bugs would be noticed and fixed already.

Copyright © 2008, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds