Hi,
My osd restart fail and here is dump messeage.
-5> 2016-04-18 09:39:13.709885 7f608462d840 15
filestore(/var/lib/ceph/osd/ceph-7) getattr
3.1a1_head/eef429a1/rb.0.130ee9.6b8b4567.0000000019b7/head//3 '_'
-4> 2016-04-18 09:39:13.709928 7f608462d840 10
filestore(/var/lib/ceph/osd/ceph-7) getattr
3.1a1_head/eef429a1/rb.0.130ee9.6b8b4567.0000000019b7/head//3 '_' =
252
-3> 2016-04-18 09:39:13.709937 7f608462d840 15 read_log missing
83830'1639861 (83830'1639860) modify
eef429a1/rb.0.130ee9.6b8b4567.0000000019b7/head//3 by
client.1491380.0:90267 2016-04-17 07:32:28.7
-2> 2016-04-18 09:39:13.709949 7f608462d840 15
filestore(/var/lib/ceph/osd/ceph-7) getattr
3.1a1_head/8835b1a1/rbd_header.3e01a6d9f226c/head//3 '_'
-1> 2016-04-18 09:39:13.709993 7f608462d840 10
filestore(/var/lib/ceph/osd/ceph-7) getattr
3.1a1_head/8835b1a1/rbd_header.3e01a6d9f226c/head//3 '_' = 593
0> 2016-04-18 09:39:13.712530 7f608462d840 -1 osd/PGLog.cc: In
function 'static void PGLog::read_log(ObjectStore*, coll_t, coll_t,
ghobject_t, const pg_info_t&, std::map<eversion_t, hobject_t>&, PGLog:
osd/PGLog.cc: 979: FAILED assert(oi.version == i->first)
ceph version 0.94.6-3-g6d41ecd (6d41ecd419e034002e5b46c541d663308c16e5e9)
1: (ceph::__ceph_assert_fail(char const*, char const*, int, char
const*)+0x72) [0xcec4f2]
2: (PGLog::read_log(ObjectStore*, coll_t, coll_t, ghobject_t,
pg_info_t const&, std::map<eversion_t, hobject_t,
std::less<eversion_t>, std::allocator<std::pair<eversion_t const,
hobject_t> > >&, PGLog::Ind
3: (PGLog::read_log(ObjectStore*, coll_t, coll_t, ghobject_t,
pg_info_t const&, std::basic_ostringstream<char,
std::char_traits<char>, std::allocator<char> >&)+0xdf) [0x93ba6f]
4: (PG::read_state(ObjectStore*, ceph::buffer::list&)+0x127) [0x923987]
5: (OSD::load_pgs()+0x8bb) [0x7f733b]
6: (OSD::init()+0xdac) [0x7fbb2c]
7: (main()+0x253e) [0x79e12e]
8: (__libc_start_main()+0xf5) [0x7f6081c6e995]
9: /usr/bin/ceph-osd() [0x7a41d7]
NOTE: a copy of the executable, or `objdump -rdS <executable>` is
needed to interpret this.
from the core message, i checked PGLog::read_log, it entered this
code. can anyone tell me which situation will cause such a problem?
and how can i repair this osd or just remove it? Thanks.
for (map<eversion_t, hobject_t>::reverse_iterator i =
divergent_priors.rbegin();
i != divergent_priors.rend();
++i) {
if (i->first <= info.last_complete) break;
if (i->second > info.last_backfill) continue;
if (did.count(i->second)) continue;
did.insert(i->second);
bufferlist bv;
int r = store->getattr(
pg_coll,
ghobject_t(i->second, ghobject_t::NO_GEN, info.pgid.shard),
OI_ATTR,
bv);
if (r >= 0) {
object_info_t oi(bv);
/**
* 1) we see this entry in the divergent priors mapping
* 2) we didn't see an entry for this object in the log
*
* From 1 & 2 we know that either the object does not exist
* or it is at the version specified in the divergent_priors
* map since the object would have been deleted atomically
* with the addition of the divergent_priors entry, an older
* version would not have been recovered, and a newer version
* would show up in the log above.
*/
assert(oi.version == i->first);
} else {
dout(15) << "read_log missing " << *i << dendl;
missing.add(i->second, i->first, eversion_t());
}
}
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
