NI3: The Net Result of Imagination, Innovation, and Investment
Friday, February 27, 2004
Yet another date related bug in EMC's software; the second one in two months. This one is not as critical as the last that brought down thousands of production Celerra systems. However, this one will completely disable your ability to monitor your environment. This bug affects EMC's ControlCenter. Depending on how you use ControlCenter to manager you entire EMC environment, you will not be notified if problems within your environment after 12:00am on 2/29/04 if you do not execute this procedure. This is a work around for the default settings that appear to be affected. Note that EMC has not released a patch for this bug and not further information is available at this time. I think a review of CS201 is in order for all programmers at EMC.

Service Alert 431 - CC 5.X.X - WARNING - On 2/29/04 at 12:00a.m. ControlCenter users will experience an issue where their ECC agent Data Collection Policies and Alert Definitions will stop being scheduled and will not run. - OPT 176542

Symptom: On 2/29/04 at 12:00 a.m. ControlCenter users will experience an issue where their ECC agent Data Collection Policies and Alert Definitions will stop being scheduled and will not run.

Fix: As a proactive measure ControlCenter users can immediately execute the following procedure from their Console to avoid this issue and not experience any service interruption on 2/29/04. If this procedure is not run prior to 2/29/04 12:00 a.m. users will continue to experience the issue until they have executed the procedure at which time normal DCP and Alert Definition scheduling and execution will be restored. For each Schedule in the system located under the Administration-> Schedules folder in the Console do the following. Right click and select the "Edit..." menu item which will start the Schedule Definition edit dialog. Within the edit dialog select Jan / 1 in the From - Month/Day selections and Dec / 31 in the To - Month/Day selections. Select OK to save the Schedule definition changes.

Again, if this procedure is executed prior to 2/29/04 users will not experience any DCP or Alert Def. downtime. If it is executed after that date then upon completion normal DCP and Alert Def. operation will be restored according to the defined schedules.
 
Thursday, February 26, 2004
A few years ago when the idea of storage resource management software first began to be discussed, few people were surprised when a number of companies jumped onto the SRM bandwagon. These vendors wanted to carve out a place for themselves as "early movers" in this new field but, alas, what most of them produced was not storage management software at all.  Rather, they provided us with software that monitored rather than managed.   [Network World on Storage]

This is so very true. Management and reporting are still very manual, even on purported industry leader's equipment and software. I had a report run for me this week to show me the number of Symmetrix arrays in my data center, the total storage they contain, and how much was storage was allocated. I had another report ran to show me how much storage was allocated to each host. Presumably the sum of the storage per host should equal the total allocated storage across all the arrays. It didn't. Our storage engineers are now off trying to figure out why.
 

Technorati Profile

Creative Commons License
This work is licensed under a Creative Commons License.