New answers tagged

1

mdraid support for discard on RAID 5 wasn't available in the 2.6 kernel series; it was added in the 3.7 kernel. I'm not aware of Red Hat backporting this functionality. Once you get a kernel with this functionality, you may find that it is disabled, because some (older) SSDs don't actually implement TRIM properly. In this case, you need to set the kernel ...


1

I wrote an article at Thinkmate that tries to give a nice overview of NVMe and works as a nice little guide on choosing the right drive and system, something that I've found to be missing online. We mostly sell Supermicro servers, and I agree - Things can get a bit confusing... That is why I wrote the article! As for adoption, I can't speak for the industry ...


0

I use RAID1 with SSD's from about an year. There are no problems there, but having different drives sometimes may kill write performance. There were some option - "writemostly" - check about it. I use this option on one server where I do raid with sata and ram disk.


0

There is no such way in any of the existing normal SSDs, you can minimize to zero the effect of garbage collection if you write only sequentially but other than that you have no control over the garbage collection. There are talks about some SSDs with features that allow the host to hint to the SSD about overall intentions but these are just hints and the ...


0

The issue with any system that gets too full is to find the next free space to write into. ZFS and any system that is copy-on-write can be especially susceptible to this issue as well as any log based system where there is a background process to clear the now-unused data after it was overwritten in principle but the new data was actually written somewhere ...


2

You would use the cheap HP SATA SSDs for boot. They're not fast. But they work just fine for the purpose. (It's clearly an Intel SATA drive) SATA is fine in this case because there's no need for dual-port SAS disks in an internal server backplane (no failover or multipath needs).


1

I had the same problem. After some googling I decided to suspend the machine remotely: CentOS6# pm-suspend and then I called a person on site to turn it on. That did the job. Yes it require some physical access but not necessarily you nor someone technical. Just call some random person around the box to press the power button.


0

It seems that discard_swap_cluster is only called from scan_swap_map which in turn is called from get_swap_page or get_swap_page_of_type. So if I'm correct, the discarding only happens when a new swap page is going to be allocated, not when a page is freed.



Top 50 recent answers are included