Changes between Version 1 and Version 2 of Production_Cluster_Status

Timestamp:

Oct 8, 2009, 10:03:50 AM (17 years ago)

Author:

eugene

Comment:

--

Production_Cluster_Status

@@ -1 @@
-<pre>
+== IPP MHPCC Production Cluster Status ==
 
-Status as of 2008-11-14
+(Up to [wiki:IPP_for_PS1 IPP for PS1])
 
-Nodes down:
+'''Status as of 2008-11-14'''
+
+=== Nodes down ===
 
 None.
 
-Notes:
+=== Notes === 
 
-- ipp015 is rebuilding disk #16
+ * ipp015 is rebuilding disk \#16
 
-Known issues:
+=== Known issues === 
 
-- random system crashes under heavy load of nodes, occasionally with a printk() of "do_IRQ: X.XXX" which appears to be caused by a hardware interrupt that does not have a handler
+ * random system crashes under heavy load of nodes, occasionally with a printk() of "do_IRQ: X.XXX" which appears to be caused by a hardware interrupt that does not have a handler
   (device driver) for it.
-- ipp004 - disk bay #12 dead (system is usable)
-- ipp005 - seems to have random instability issues beyond the do_IRQ issue of the other nodes
-- ipp008 - disk bay #7 dead (system is usable)
-- ipp013 & ipp016 have dead fans but it is not the fans themselves (tried replacing them).  It appears to be the cabling that leads up to the fan modules that's gone bad.
-- forcedeth.c max_interrupt_work is still too conservative (requires a cluster restart to change).
-- ipp025 - suspect bad memory module (system is usable)
-- Our contact at MHPCC, Brad Thomas, notified me that an alarm from cabinet 4 managed power strip had been set off last evening due to high loads on nodes (ipp017, 030-036).
+ * ipp004 - disk bay #12 dead (system is usable)
+ * ipp005 - seems to have random instability issues beyond the do_IRQ issue of the other nodes
+ * ipp008 - disk bay #7 dead (system is usable)
+ * ipp013 & ipp016 have dead fans but it is not the fans themselves (tried replacing them).  It appears to be the cabling that leads up to the fan modules that's gone bad.
+ * forcedeth.c max_interrupt_work is still too conservative (requires a cluster restart to change).
+ * ipp025 - suspect bad memory module (system is usable)
+ * Our contact at MHPCC, Brad Thomas, notified me that an alarm from cabinet 4 managed power strip had been set off last evening due to high loads on nodes (ipp017, 030-036).
   Please consider distributing your jobs evenly across ipp production nodes.
 
-  Below is the layout at MHPCC.
+===  Layout at MHPCC ===
+{{{
   cabinet0: ipp020-21
   cabinet1: ipp012-014,008,016,018-019
@@ -30 +33 @@
   cabinet3: ipp023-029
   cabinet4: ipp030-036,017
-
-http://kiawe.ifa.hawaii.edu/IPPwiki/index.php/Production_Cluster_Status
-
-</pre>
+}}}