Messages in this thread | | | Date | Fri, 17 Sep 1999 09:14:47 -0500 | From | Tim Walberg <> | Subject | Re: NFS problem: __nfs_fhget: inode still busy |
| |
On 09/17/1999 10:18 +0100, Jakma, Paul wrote: >> Hi Trond, >> >> Yesterday i started getting the famous "access beyond end of block device" >> errors on the nfs server. I ran an fsck on all drives, and there were a lot >> of errors. Once the server was back up the client booted fine again. >> >> I don't understand what nfs stale/busy inode errors are about, so can i put >> those errors down to corrupted filesystems on the server? (though they >> occurred a day before "access beyond end of block device" on the server.) >>
No. Stale NFS file handles result from a file being removed either on the server or on a different client, while the client you are dealing with had the file open in some way (usually happens more with directories than files in my experience - someone is 'cd'ed into a directory on the client when another client or the server removes that directory). Without perusing the source code, I don't recall the exact events that cause busy inode errors, but I suspect they are fairly similar. Neither one is associated with corrupt file systems per se, although a corrupt file system may cause files to disappear unexpectedly, which can definitely cause these symptoms.
-- +------------------------------+--------------------------+ | Tim Walberg | Phone: 847-782-2472 | | TERAbridge Technologies Corp | FAX: 847-623-1717 | | 1375 Tri-State Parkway | twalberg@terabridge.com | | Gurnee, IL 60031 | 800-SKY-TEL2 PIN 9353299 | +------------------------------+--------------------------+ [unhandled content-type:application/pgp-signature] | |