From Martin's valgrind log, it appears it's the dreaded memory problem in imap_set_batch(), which caused another bugs, and which has been fixed in 2.1.0. (or sooner. I'm not sure).
==18390== Address 0x5C2ECE8 is 232 bytes inside a block of size 504 free'd ==18390== at 0x401D048: free (vg_replace_malloc.c:235) ==18390== by 0x47F015B: g_free (in /usr/lib/libglib-2.0.so.0.1000.2) ==18390== by 0x80DD2EF: (within /usr/bin/sylpheed-claws-gtk2) ==18390== by 0x80C93C8: folder_item_set_batch (in /usr/bin/sylpheed-claws-gtk2) -- Colin
