The plot thickens...
Looking at the next relation I see a similar symptom of a single valid
block at the end of several segments of nuls. This relation is also a
btree on the same table and has a header in the near vicinity of the
xlog:
d9de7pcqls4ib6=# select
(page_header(get_raw_page('event_data_event_id_person_id', 'main',
3631161))).* ; lsn | tli | flags | lower | upper | special | pagesize |
version | prune_xid
------------+-----+-------+-------+-------+---------+----------+---------+-----------EA1/63A0A8 | 6 | 0 | 1180 |
3552 | 8176 | 8192 |
4 | 0
(1 row)
And indeed it's the very next xlog record:
[cur:EA1/638988, xid:1418089147, rmid:11(Btree), len/tot_len:18/5894,
info:8, prev:EA1/637140] insert_leaf: s/d/r:1663/16385/1364767 tid
2746914/219
[cur:EA1/638988, xid:1418089147, rmid:11(Btree), len/tot_len:18/5894,
info:8, prev:EA1/637140] bkpblock[1]: s/d/r:1663/16385/1364767
blk:2746914 hole_off/len:1180/2372
[cur:EA1/63A0A8, xid:1418089147, rmid:1(Transaction),
len/tot_len:32/64, info:0, prev:EA1/638988] d/s:16385/1663 commit at
2014-01-21 05:41:11 UTC