Score:0

RAID drive - Mechanical Positioning Error

in flag

We've just had a RAID drive go offline. There were no log entries indicating failure prior to 'Mechanical positioning error' - six 'Peripheral device write fault' entries were logged after, and before the controller took it offline and started to rebuild with the hot spare.

There seems very little information about this particular error, apart from a couple of conflicting articles from Sun (this is not a Sun server) one indicates it's just a transient error, another that it can bring the entire controller down.

My interpretation would be that the drive head seized and the drive is toast - and if the RAID controller recognised it and took it offline then I am lucky. Would this be a reasonable assessment?

The failed drive was one that was replaced in 2018, one of many (11 to date) Seagate drives that have failed on this array.

Log entries

Score:1
ru flag

Drives can fail after some time, that's why you're using RAID. Just replace it.

in flag
Yes, it was replaced - with the ten before it. The question, however, was about the fault. Usually there are a trail of predictive failure errors leading up to one falling over. The question was 'what is a mechanical positioning error?'
Zac67 avatar
ru flag
By my experience, about 30% of drives go without any previous signs of coming failure. A mechanical positioning error is likely a head failure. Does it *really* matter? HDDs are consumables.
in flag
In the overall scheme of things it matters not - a dead drive is a dead drive BUT I do like to record the reasons for the failures - a post-mortem if you will - to get a better idea of how drives perform in our environment. There is no detailed explanation that I can find online. My guess would be a head seizing, I was kinda looking for some confirmation or otherwise. In my experience drives start with read-write errors before they fall over and not close to 30% immediate collapse.
mangohost

Post an answer

Most people don’t grasp that asking a lot of questions unlocks learning and improves interpersonal bonding. In Alison’s studies, for example, though people could accurately recall how many questions had been asked in their conversations, they didn’t intuit the link between questions and liking. Across four studies, in which participants were engaged in conversations themselves or read transcripts of others’ conversations, people tended not to realize that question asking would influence—or had influenced—the level of amity between the conversationalists.