Hi, Am having a heck of a time trying to figure out a bottleneck for NFS transfers. I have 2 Mandy boxes, acting only as NFS file servers.
BOX1: 1.8Ghz, Mandt 8.2 Kernel 2.4.18, reiserfs, NFSv3 on 100MbitFD [ NIC:Intel Corp. 82820 (ICH2) Chipset Ethernet Controller (rev 03) ] BOX2: 2.4Ghz Dual Xeon Processors, Mandy 9.0 Kernel 2.4.20, LVM is being used, file system is still reiserfs, NFSv3 on 100MbitFD [ NIC:Intel Corp. 82557/8/9 [Ethernet Pro 100] (rev 0d) ] -Both run iptables, no effect on transfer rates when iptables is on/off on either. Rate of transfer differs significantly!! They both serve disks (1Tb) to a single tru64 Unix server (Using IDE Raid, same configuration of RAIDcard and storage box) a 77GB transfer from Tru64Unix disk to an NFS mounted disk from BOX1, took about 8hrs. A 77GB transfer from Tru64Unix disk to an NFS mounted disk from BOX2, has taken two days and counting!! No NFS errors reported. Everything seems to be working perfectly!, except for the transfer rate! Even utlizing the RAID array without LVM, but on BOX2, yields the same very slow transfer rates! Is it something with Mandy9.0? or what else could help in troubleshooting the issue here? Clueless Richard, _Thanks ===== __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com
Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com