Recent content by Patrick

For folks who have not been following, Sam has been working in the studio and shooting videos for us, like the Dell AI Factory video. This series started because Sam had a question on why we had so many T7 Shield SSDs versus others, as we were loading the footage onto one from Dell's...

Well, I was a bit shocked to see myself in that one so much. Also, I am still under NDA for what I saw that we did not show, so I cannot go into any of that. We also only did one video of the cluster. A few key notes: What I was confused about is that they often they were saying generators...

Hey @d4ve @TRACKER and @Koop Yesterday was a day. I had a 4 hr flight delay for a cockpit switch cover, had to swap flights to make it on a plane to Taipei for Computex... and then as I get on the plane I get a red alert SOS from Dell saying a 3rd party had an issue and we had to pull it. I...

That is exactly what happened. BTW we have the MS-A2 so expect a review coming.

The answer behind this is a memory transaction, which is gobbling up GBs. It has taken a long time to find not just what was consuming memory, but also the "why" behind it. I think we have a fix that we will try this weekend. If not, there is another window next week to make the change. From a...

Thanks! A long way indeed. TBH these days it is a lot of changing diapers and bottles.

Thanks all. I just did my first full night shift and am pretty tired. The little guy and I “watched” the Monaco GP this morning. Hopefully we get past the shock and back to full speed on STH later this week.

I am not sure how to, but they come from the units that have GPUs and those that have 65W TDP CPUs.

So the captcha seems to be the problem. The challenge is that the amount of spam is wild these days and it does help cut it down quite a bit.

It was a power freak-out in the chassis. I figured there was a forum thread on them. The funny part @jamesthetechie is that the C6100 was a super high-volume line before many of the the hyper-scalers went OCP.

This should be the first message on the forums on the new cluster. Net downtime was under 3 minutes.

It is just easier to do this as a straight transfer. I am hoping it takes way less than an hour.

I am hoping this will only be for a maximum of 1 hour. Here is the issue: We need the forums to get from an old cluster to a new cluster and it is going to have to be a cold migration. My plan is either Friday night or Saturday night when activity is lower. The good news is that it will have...

@piranha32 should be fixed now. Thanks. We had to re-enable reCapthca. Our Akismet API requests have burned through two paid license levels this month.

That was the worst! Super over-provisioned Kingston "data center" SSDs did not have power inrush protection so a surge in the Dell C6100 took out the drives in every node. People wonder why I have become a fan of losing performance by using mixed brands of drives in RAID 1 in the hosting...