Windows logs these events using volume IDs which don't exactly match with the physical drive or drive letters very well. Either due to it being a cheap and slow USB drive, or on systems that have flaky USB chipsets. The other issue is that USB drives are notorious for throwing bad block errors. This cuts down the fast positives immensely. We set our RMM to alert only if 5 or more events happen within one hour. The only downside to monitoring this is that it is a little noisy. I have successfully done this with Dell, HP and Lenovo. It is a little tricky getting OEMs to warranty these drives as their normal SMART tests show fine, but when you tell them that you are monitoring this on 2k other endpoints without issue, they will eventually be able to get approval to warranty replace the drive. Since we move to monitoring bad blocks, I believe not a single client has lost data.
The bad block errors can proceed a drive failing by weeks or months. These alerts will never be present on a drive functioning properly, but will show up on a drive that is starting to fail. We stopped monitoring for SMART errors entirely and instead monitor the system event log for bad block alerts (event id 7). Now the question is: Did it happen? Did I miss the "to go" tool for SSD testing and monitoring or is there really still not much available? When I first "investigated" this topic a couple of years ago, everybody was like "oh well, there isn't really anything out for SSDs yet but we expect this to happen soon". Also different tools that come directly from the manufacturers of the devices (like Micron Storage Executive for example) didn't really impress me so far. is of course still around on SSDs but I feel that it is by far not as useful on SSDs as it is on HDDs. This raises the question: How do we test and monitor SSDs? However in 2021 HDDs can more or less be considered legacy technology, at least when it comes to the PCs and notebooks of a typical MSP client. is obviously a quite handy thing when it comes to HDDs and most of us have quite extensive experience in making use of it. Most of you (at least if they did not just start very recently) probably have their tools and methods to test and monitor HDDs.
0 kommentar(er)
