diff options
| author | Trond Myklebust <trond.myklebust@hammerspace.com> | 2020-08-11 13:36:32 -0400 |
|---|---|---|
| committer | Trond Myklebust <trond.myklebust@hammerspace.com> | 2020-08-12 11:20:29 -0400 |
| commit | 563c53e73b8b6ec842828736f77e633f7b0911e9 (patch) | |
| tree | 2fc1fe8cdc849167b3cd0632ca799586d823cce6 /fs/nfs/pnfs.c | |
| parent | a5032910c5e542dddedeaa6500e5eab3f1175384 (diff) | |
NFS: Fix flexfiles read failover
The current mirrored read failover code is correctly resetting the mirror
index between failed reads, however it is not able to actually flip the
RPC call over to the next RPC client.
The end result is that we keep resending the RPC call to the same client
over and over.
The fix is to use the pnfs_read_resend_pnfs() mechanism to schedule a
new RPC call, but we need to add the ability to pass in a mirror
index so that we always retry the next mirror in the list.
Fixes: 166bd5b889ac ("pNFS/flexfiles: Fix layoutstats handling during read failovers")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Diffstat (limited to 'fs/nfs/pnfs.c')
| -rw-r--r-- | fs/nfs/pnfs.c | 4 |
1 files changed, 3 insertions, 1 deletions
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index b5baf36d4de5..40332c758d84 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -2939,7 +2939,8 @@ pnfs_try_to_read_data(struct nfs_pgio_header *hdr, } /* Resend all requests through pnfs. */ -void pnfs_read_resend_pnfs(struct nfs_pgio_header *hdr) +void pnfs_read_resend_pnfs(struct nfs_pgio_header *hdr, + unsigned int mirror_idx) { struct nfs_pageio_descriptor pgio; @@ -2950,6 +2951,7 @@ void pnfs_read_resend_pnfs(struct nfs_pgio_header *hdr) nfs_pageio_init_read(&pgio, hdr->inode, false, hdr->completion_ops); + pgio.pg_mirror_idx = mirror_idx; hdr->task.tk_status = nfs_pageio_resend(&pgio, hdr); } } |