Segfault after BlockchainLMDB::do_resize

[pmknutsen]

bitmonerod silently fails when launched after executing BlockchainLMDB::do_resize. I have reset the database, and re-synced with the same result. The following is the last output when running the deamon. Box runs Ubuntu 14.04 LTS, with 8GB RAM and tons of HDD space to spare:

2015-Sep-11 22:47:38.748646 BlockchainLMDB::need_resize
2015-Sep-11 22:47:38.748851 DB map size:     1073741824
2015-Sep-11 22:47:38.749055 Space used:      967413760
2015-Sep-11 22:47:38.749251 Space remaining: 106328064
2015-Sep-11 22:47:38.749446 Size threshold:  0
2015-Sep-11 22:47:38.749818 Percent used: 0.9010  Percent threshold: 0.8000
2015-Sep-11 22:47:38.751299 Threshold met (percent-based)
2015-Sep-11 22:47:38.751443 LMDB memory map needs resized, doing that now.
2015-Sep-11 22:47:38.751589 BlockchainLMDB::do_resize
2015-Sep-11 22:47:38.751828 LMDB Mapsize increased.  Old: 1024MiB, New: 2048MiB

Segfault as reported in /var/log/syslog

kernel: [859428.738710] bitmonerod[8170]: segfault at 72a0001c ip 083ec820 sp bf814980 error 4 in bitmonerod[8048000+58e000]
kernel: [859502.299108] bitmonerod[8180]: segfault at 72a0001c ip 083ec820 sp bf8658e0 error 4 in bitmonerod[8048000+58e000]
kernel: [859520.137126] bitmonerod[8188]: segfault at 72a0001c ip 083ec820 sp bfa2fb90 error 4 in bitmonerod[8048000+58e000]

[pmknutsen]

Found the cause of this segfault. Bitmonero was running on 32-bit Ubuntu. Problem resolved when updating OS to 64 bit.