We are seeing allot of messages like the following: Lustre: 3177:0:(router.c:167:lnet_notify()) Ignoring prediction from 141.212.30.181 at tcp of 141.212.31.123 at tcp down 9786657437 seconds in the future Lustre: 3243:0:(ldlm_lib.c:519:target_handle_reconnect()) nobackup- OST0000: 606834a0-0818-74c2-6847-efb3f1628347 reconnecting Lustre: 3243:0:(ldlm_lib.c:519:target_handle_reconnect()) Skipped 1 previous similar message Lustre: 3243:0:(ldlm_lib.c:519:target_handle_reconnect()) nobackup- OST0001: 606834a0-0818-74c2-6847-efb3f1628347 reconnecting Whats this prediction of host failing? Brock Palen Center for Advanced Computing brockp at umich.edu (734)936-1985
On Wed, Jan 23, 2008 at 10:06:54AM -0500, Brock Palen wrote:> We are seeing allot of messages like the following: > > Lustre: 3177:0:(router.c:167:lnet_notify()) Ignoring prediction from > 141.212.30.181 at tcp of 141.212.31.123 at tcp down 9786657437 seconds in > the futureIt''s a known & fixed bug: https://bugzilla.lustre.org/show_bug.cgi?id=14300 The fix will be in Lustre 1.4.12 and 1.6.5. Isaac