LSI 9266/9271 as well in an affected range unless ECO’d
> On Apr 19, 2023, at 3:13 PM, Sebastian <sebci...@gazeta.pl> wrote: > > I want add one thing to what other says, we discussed this between > Cephalocon sessions, avoid HP controllers p210/420, or upgrade firmware to > latest. > These controllers has strange bug, during high workload they restart itself. > > BR, > Sebastian > >> On 19 Apr 2023, at 08:39, Janne Johansson <icepic...@gmail.com> wrote: >> >> Den ons 19 apr. 2023 kl 00:55 skrev Murilo Morais <mur...@evocorp.com.br>: >>> Good evening everyone! >>> Guys, about the P420 RAID controller, I have a question about the operation >>> mode: What would be better: HBA or RAID-0 with BBU (active write cache)? >> >> As already said, always give ceph (and zfs and btrfs..) the raw disks >> to handle for itself, instead of doing hardware striping/raiding and >> so on. >> >> There are multiple reasons for this, including weird/bad firmware and >> (for raid0) loss of redundancy and so on, but there are also corner >> cases where you may need/want to move disks from one box to another. >> If you use "raw" disks, then what is on the platters most often is 1:1 >> to what the computer sees, and hence if you move the drive over to >> another box, it will not matter if it has a HBA or Raid card, what >> brand of disk controller it is and so on. The new server will also see >> 1:1 of the data the former machine wrote and this is what you want. >> >> If you have a raid0/jbod/raid1/raidX setup on a particular raid >> controller card with a particular firmware version, you may not be >> able to move your complete raid set over to a newer box. Perhaps the >> raid card model is no longer available, perhaps the raid firmware >> logic is new and different on how to stripe data or place sectors on >> the raid members. Perhaps it detects disks in the wrong order or >> something. >> >> It is very possible that you have zero problems with raid setups, but >> the extra complexities when moving raid sets from one generation of >> computer/controller to another makes the risk of some kind of >> portability issue non-zero. Old grumpy storage admins don't like >> non-zero risks if we actually can avoid it. >> >> -- >> May the most significant bit of your life be positive. >> _______________________________________________ >> ceph-users mailing list -- ceph-users@ceph.io >> To unsubscribe send an email to ceph-users-le...@ceph.io > _______________________________________________ > ceph-users mailing list -- ceph-users@ceph.io > To unsubscribe send an email to ceph-users-le...@ceph.io _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-le...@ceph.io