libceph: don't go through with the mapping if the PG is too wide
With EC overwrites maturing, the kernel client will be getting exposed
to potentially very wide EC pools. While "min(pi->size, X)" works fine
when the cluster is stable and happy, truncating OSD sets interferes
with resend logic (ceph_is_new_interval(), etc). Abort the mapping if
the pool is too wide, assigning the request to the homeless session.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Reviewed-by: Sage Weil <sage@redhat.com>
diff --git a/net/ceph/osdmap.c b/net/ceph/osdmap.c
index 2374956..6824c0e 100644
--- a/net/ceph/osdmap.c
+++ b/net/ceph/osdmap.c
@@ -2013,8 +2013,14 @@ static void pg_to_raw_osds(struct ceph_osdmap *osdmap,
return;
}
- len = do_crush(osdmap, ruleno, pps, raw->osds,
- min_t(int, pi->size, ARRAY_SIZE(raw->osds)),
+ if (pi->size > ARRAY_SIZE(raw->osds)) {
+ pr_err_ratelimited("pool %lld ruleset %d type %d too wide: size %d > %zu\n",
+ pi->id, pi->crush_ruleset, pi->type, pi->size,
+ ARRAY_SIZE(raw->osds));
+ return;
+ }
+
+ len = do_crush(osdmap, ruleno, pps, raw->osds, pi->size,
osdmap->osd_weight, osdmap->max_osd);
if (len < 0) {
pr_err("error %d from crush rule %d: pool %lld ruleset %d type %d size %d\n",