Comments for Wire Turf http://www.wireturf.com Wed, 12 Oct 2011 06:31:05 +0000 hourly 1 http://wordpress.org/?v=3.2 Comment on In the aftermath of Intermedia’s extended outage, an important lesson to be learned for SAS providers by shawn http://www.wireturf.com/2010/03/12/in-the-aftermath-of-intermedias-extended-outage-an-important-lesson-to-be-learned/comment-page-1/#comment-127 shawn Wed, 12 Oct 2011 06:31:05 +0000 http://www.wireturf.com/?p=141#comment-127 Same here...been down since yesterday. I like how their "system status" page says that all systems are operational. They must mean by Intermedia's standards. Same here…been down since yesterday. I like how their “system status” page says that all systems are operational. They must mean by Intermedia’s standards.

]]> Comment on In the aftermath of Intermedia’s extended outage, an important lesson to be learned for SAS providers by Jim http://www.wireturf.com/2010/03/12/in-the-aftermath-of-intermedias-extended-outage-an-important-lesson-to-be-learned/comment-page-1/#comment-126 Jim Tue, 11 Oct 2011 14:55:58 +0000 http://www.wireturf.com/?p=141#comment-126 And going on 24 hours of outage for us. Not acceptable given this is the same explanation (SAN Array Failure) given last time this happened. Fool me once... And going on 24 hours of outage for us. Not acceptable given this is the same explanation (SAN Array Failure) given last time this happened. Fool me once…

]]> Comment on In the aftermath of Intermedia’s extended outage, an important lesson to be learned for SAS providers by mookie http://www.wireturf.com/2010/03/12/in-the-aftermath-of-intermedias-extended-outage-an-important-lesson-to-be-learned/comment-page-1/#comment-125 mookie Tue, 11 Oct 2011 11:18:27 +0000 http://www.wireturf.com/?p=141#comment-125 6 more hours of SAN related outage yesterday. 6 more hours of SAN related outage yesterday.

]]> Comment on First Impressions using Google+ by Josh Schneider http://www.wireturf.com/2011/07/11/first-impressions-using-google/comment-page-1/#comment-57 Josh Schneider Mon, 11 Jul 2011 13:40:21 +0000 http://www.wireturf.com/?p=223#comment-57 Hi David, long time no speak. Hope you are doing well and don't miss Razorgator too much. I would love to give Google+ a shot. Shoot me an invite if you have an extra. Hi David, long time no speak. Hope you are doing well and don’t miss Razorgator too much. I would love to give Google+ a shot. Shoot me an invite if you have an extra.

]]> Comment on In the aftermath of Intermedia’s extended outage, an important lesson to be learned for SAS providers by Joel http://www.wireturf.com/2010/03/12/in-the-aftermath-of-intermedias-extended-outage-an-important-lesson-to-be-learned/comment-page-1/#comment-55 Joel Sun, 31 Oct 2010 08:03:47 +0000 http://www.wireturf.com/?p=141#comment-55 Intermedia is DOWN now (early Sunday morning Oct 31). Intermedia is DOWN now (early Sunday morning Oct 31).

]]> Comment on In the aftermath of Intermedia’s extended outage, an important lesson to be learned for SAS providers by Chris http://www.wireturf.com/2010/03/12/in-the-aftermath-of-intermedias-extended-outage-an-important-lesson-to-be-learned/comment-page-1/#comment-52 Chris Tue, 25 May 2010 01:28:29 +0000 http://www.wireturf.com/?p=141#comment-52 maybe off topic here, but does anyone have a good solution to move from intermedia? I am tired of answering the question, what is wrong with our email now. I have installed microsoft exchange 2010 and have actually started migrating people, but the task of downloading the pst files from intermedia is no fun. maybe off topic here, but does anyone have a good solution to move from intermedia? I am tired of answering the question, what is wrong with our email now. I have installed microsoft exchange 2010 and have actually started migrating people, but the task of downloading the pst files from intermedia is no fun.

]]> Comment on In the aftermath of Intermedia’s extended outage, an important lesson to be learned for SAS providers by Bo http://www.wireturf.com/2010/03/12/in-the-aftermath-of-intermedias-extended-outage-an-important-lesson-to-be-learned/comment-page-1/#comment-51 Bo Fri, 07 May 2010 03:56:30 +0000 http://www.wireturf.com/?p=141#comment-51 Hi David, How's life? Heard you left OleOle? Is everything going well with you. Bo Hi David,

How’s life?

Heard you left OleOle? Is everything going well with you.

Bo

]]> Comment on In the aftermath of Intermedia’s extended outage, an important lesson to be learned for SAS providers by John Martin http://www.wireturf.com/2010/03/12/in-the-aftermath-of-intermedias-extended-outage-an-important-lesson-to-be-learned/comment-page-1/#comment-50 John Martin Tue, 27 Apr 2010 10:58:16 +0000 http://www.wireturf.com/?p=141#comment-50 *Disclosure I Work for an EMC Competitor* To be fair to EMC, the system in the initial failure did continue to operate but at a degraded service level. This is a characteristic of all modular storage arrays that work in high availability pairs. The level of degradation depends on the make of the array and the settings used, and it appears that Expedia took the option of choosing a configuration that put data preservation ahead of performance. Unfortunately this choice which is understandable had unpleasant and unforseen consequences. What surprises me is that the impact of failover either wasnt tested (which is really hard to do in a multi-tenant environment), or mitigated by only using half the possible performance of the array. Having said that, even when only using half the possible performance on a mid-range arrays, maintaining that performance when half the array fails requires that write caching remains on, which is a dangerous practice because if the second controller fails, cached data can be lost. If I were intermedia, I'd seriously consider looking for a SAN that is either easier to test under failure conditions, or doesnt lose the majority of its performance when one of the components fails. Given the additional outages that have happened since, I'd expect that they'd be doing that as a matter of urgency. *Disclosure I Work for an EMC Competitor*

To be fair to EMC, the system in the initial failure did continue to operate but at a degraded service level. This is a characteristic of all modular storage arrays that work in high availability pairs. The level of degradation depends on the make of the array and the settings used, and it appears that Expedia took the option of choosing a configuration that put data preservation ahead of performance. Unfortunately this choice which is understandable had unpleasant and unforseen consequences.

What surprises me is that the impact of failover either wasnt tested (which is really hard to do in a multi-tenant environment), or mitigated by only using half the possible performance of the array.

Having said that, even when only using half the possible performance on a mid-range arrays, maintaining that performance when half the array fails requires that write caching remains on, which is a dangerous practice because if the second controller fails, cached data can be lost.

If I were intermedia, I’d seriously consider looking for a SAN that is either easier to test under failure conditions, or doesnt lose the majority of its performance when one of the components fails.

Given the additional outages that have happened since, I’d expect that they’d be doing that as a matter of urgency.

]]> Comment on In the aftermath of Intermedia’s extended outage, an important lesson to be learned for SAS providers by Eric http://www.wireturf.com/2010/03/12/in-the-aftermath-of-intermedias-extended-outage-an-important-lesson-to-be-learned/comment-page-1/#comment-47 Eric Fri, 16 Apr 2010 16:49:50 +0000 http://www.wireturf.com/?p=141#comment-47 Yes, at present it's only one server cluster that's down- EXCH020 - but it is down HARD and of course that's the one we're on. They had to stop accepting inbound mail connections. 2nd day without business email. Charming. Yes, at present it’s only one server cluster that’s down- EXCH020 – but it is down HARD and of course that’s the one we’re on. They had to stop accepting inbound mail connections. 2nd day without business email. Charming.

]]> Comment on In the aftermath of Intermedia’s extended outage, an important lesson to be learned for SAS providers by David http://www.wireturf.com/2010/03/12/in-the-aftermath-of-intermedias-extended-outage-an-important-lesson-to-be-learned/comment-page-1/#comment-46 David Fri, 16 Apr 2010 15:31:46 +0000 http://www.wireturf.com/?p=141#comment-46 yikes! 2 days running is totally unacceptable. yikes! 2 days running is totally unacceptable.

]]>