----- Original Message ----- From: "James Wilkinson" <james@xxxxxxxxxxxxxxxxxxx> To: "For users of Fedora Core releases" <fedora-list@xxxxxxxxxx> Sent: Tuesday, September 28, 2004 4:40 PM Subject: Re: FC2+rsync+file corruption > Mike McMullen wrote: > > I am experiencing file+filesystem corruption when trying to do an rsync > > to an FC2 system. > > > > I have two FC2 systems. I use rsync to create a mirror of our office data > > on the one system as a backup during different times of the day. > > > > Both systems were originally running RH7.3. A week a go I freshly installed > > FC2 from CD and updated to the most current version of packages on the > > main server and moved our applications over. The system ran fine for the > > week. So I installed FC2 from CD and updated packages on the backup > > server this weekend. This all went great. > > > > Today as I am trying to sync the backup to the main, I get the following > > errors after the rsync goes merrily on for about 8 gigabytes synced to the > > backup server. > > > > failed to set times on "somefilenamehere": Read-only file system > > > > Then I get the following error on every file after that: > > > > mkstemp "/EPC/data/SuppDocs/Vista/MooreM-1rst/.anotherfilename.Wnopm8" failed: Read-only file system > > > > The entire file system which is "/" is now read-only. If I try to reboot I get that the filesystem needs > > checked and drops me into maintenance. I fsck the partition and it find all kinds of problems > > and some files are mangled. > > Sounds weird. > > Run uname -a, please: there's an outside chance that you might have got > a particular test kernel (2.6.8-1.538) that has been causing problems > just like these. > > Otherwise, it *could* always be a case of recently-developed memory > problems: run memtest86 to check. > > Run smartctl -l error /dev/hdx on the underlying hard drives. > > I'm not going to be popular for asking this, but is the problem > repeatable? Does it happen with a kernel.org kernel? > > If you get to this point and the errors are repeatable, please enter > the problem into bugzilla.redhat.com. > > James. Hi James! Thanks for the reply! It happened two times on that server. I had another server handy and went through the install process repeating every thing on it and nothing happened. It worked as expected. I was under a time crunch and took the path of least resistance. The kernel is 2.6.8-1.521. I am suspecting like you a recent memory problem. That server is at another location and I will be back there tomorrow. I plan on running memtest86 on it. Thanks! Mike
