Re: NFS, intermittent 'RPC struct is bad' errors
- In reply to: Rick Macklem : "Re: NFS, intermittent 'RPC struct is bad' errors"
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Tue, 20 Aug 2024 15:28:45 UTC
On Wed, Jun 19, 2024 at 10:05 AM Rick Macklem <rick.macklem@gmail.com> wrote: > On Tue, Jun 18, 2024 at 11:32 PM Lexi Winter <lexi@le-fay.org> wrote: > > i have a few systems running NFSv4 on FreeBSD, using Kerberos (MIT > > Kerberos KDC), with the server exporting ZFS filesystems. > > > > recently i've noticed intermittent errors of 'RPC struct is bad' when > > writing to the NFS server, which usually resolves itself after retrying. > > for example: > [...] > No one else has reported anything like this recently, We are also seeing intermittent "RPC struct is bad" from FreeBSD NFSv4.2 clients accessing ZFS filesystems. There are a few differences between our situation and that reported by Lexi Winter: - NFS servers are Debian 12. - We do not use kerberos, noncontigwr, or delegations. - It does not resolve itself after retrying. In our case, it seems to "infect" directories as viewed from a certain FreeBSD client machine after attempting to modify (usually delete) that directory. once a directory is affected, it can no longer be viewed or removed from that client. Other clients (FreeBSD or otherwise) are not affected. It doesn't seem to be cached or anything. I.e., it doesn't go away if you leave that directory alone for a while (whether a few minutes, few hours, or a full day). Once the directory is removed from another (identical) client, everything is fine. It seems to happen randomly, on the order of once every few billion NFS calls, so whatever it is, it's a very, very edge case. ZFS+NFS seems to have a few of those, and I'm sure cross-platform mounts aren't helping. Thanks!