Wed, 07/13/2011 - 12:06 — dcwaterm.ncsu.edu
Started a workaround section today on the xteams website. First workaround added for LDAP client issues on Unix. This might help anyone after hours and oncall for Unix servers experiencing an LDAP connection problem. The issue has been observed several times on older Solaris 10 update 2. Never is it a good time for this bug (especially like recent event at 2 AM last Sunday morning). At any rate, try this workaround if warranted. Might save some of us a lot of time and avoid a full reboot. The LDAP client workaround has saved us on three occasions ov
Tue, 06/07/2011 - 13:01 — dcwaterm.ncsu.edu
Current T3 systems use ILOM 3.0 or higher. Which no longer supports the legacy ALOM commands. That's a change some of us will have to get use to as the command syntax is very different. I found a table comparing the old and new commands that might prove helpful.
For help on ALOM to ILOM command syntax, click this link: ALOM-to-ILOM
Tue, 11/02/2010 - 13:06 — dcwaterm.ncsu.edu
Occasionally we receive a report of a Solaris server with a swap space shortage. Much of the time there is no problem. We get enough of these reports that I've set up a technical doc Solaris Memory Management (going beyond swap -s) on this wildly misunderstood subject of memory management and swap disk space. It's a good start on the Solaris VM subsystem. I expect this doc to be updated and broadened in time.
Tue, 09/14/2010 - 19:44 — dcwaterm.ncsu.edu
If using server virtualization such as in this case of Solaris containers and zones, we find advantages provided in the tooling for migrating virtualized servers from one host to another. Traditionally, before we ever had server virtualization, the moving of servers to new or different hardware often required a lot of time and effort. However with Solaris 10 and the zone migration features, we were recently delighted in finding the move of virtualized container zones much faster and easier than expected. As long as the source and target host servers are of equal spec
Tue, 04/13/2010 - 20:40 — dcwaterm.ncsu.edu
One issue with sharing NICs when more than one VLAN is defined on a Solaris host with non-global zones is routing between non-global zones for the same physical host. By default routing is direct, not across the wire, regardless if the other non-global zones are in different VLANs. Such is usually deemed a compromise, espeically when there is an intended hardware Firewall for security residing between two wired VLANS (as in our case). read more updated information
Thu, 04/01/2010 - 17:03 — dcwaterm.ncsu.edu
After a domain crash on the E25k, normally the domain will restart. Such as the case earlier this week with the oradev1 domain on the DC1 E25k. Problem determination uncovered one CPU experienced a corrupted ecache location which when detected immediately causes a system panic crash. Diagnostics for root cause normally are performed in the hours well past the system recovers. In this case the system had been restarted and running 24 hours by the time of pinpointing root cause. Protocol is to replace such failing components rather than risk a recurrence.