> Hi all! > > We are running cyrus-imapd with all files (mailboxes, Berkeley-DB, > mmap()ed files) on an ext3 filesystem (with data=ordered) on a iSCSI > block device (and the actual storage is on a NetApp box if that matters) > on a Xeon 3.2GHz CPU (with 4 cores/2 HT thread - ). > The kernel is 2.6.18-128.el5 (from the RHEL5/CentOS-5 updates). > > On a low-(to-now-)volume test cluster, everything runs fine. On the > high-volume production cluster[0], we experienced a corrupted BDB. > Googling didn't yield any useful result on the cause (or hint on > improvement of the configuration). > > Does anyone have experience with such a setup? > Is it possible to (simply) move the __* files to a local drive (or a > tmpfs/ramfs/ramdisk)? > > Any hints or questions or feedback or ... is greatly appreciated, Hi, As a first step I suggest to get rid of the Berkeley-DB files and convert them to skiplist. Regards, Simon ---- Cyrus Home Page: http://cyrusimap.web.cmu.edu/ Cyrus Wiki/FAQ: http://cyrusimap.web.cmu.edu/twiki List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html
