Arjen van der Meijden
2005-Jun-24 08:49 UTC
[Xapian-discuss] Quartzcompact in svn-6307 fails
Hi List, I'm trying to test some of Olly's changes to the compaction of quartz, but appareantly quartzcompact just doesn't work at all, or of course my installation is bugged. I've installed svn-6703 on a gentoo system which is quite up-to-date in terms of gcc (3.3)/glibc versions. When I do quartzcompact on our (with 0.8.4) compacted database without any parameters it just says 'aborted' right after it started, with the -n and -F parameters the same. But with -F and -n combined, it segfaults with this backtrace: #0 0xb7cf6003 in memmove () from /lib/libc.so.6 #1 0xb7ee0cdc in Btree::compact (this=0xbfffe780, p=0x8057fa8 "\004\026?\026?\026?\026?\026?\237") at btree.h:181 #2 0xb7ee15f0 in Btree::add_item (this=0xbfffe780, j=0) at btree.cc:832 #3 0xb7ee1b7f in Btree::add_kt (this=0xbfffe780, found=false) at btree.cc:998 #4 0xb7ee222f in Btree::add (this=0xbfffe780, key=@0xbfffe6b0, tag {static npos = 4294967295, _M_dataplus = {<std::allocator<char>> = {<No data fields>}, _M_p = 0x9933604 "0??s?g\031\006???\237?]~k?\225%!?\022?\202[gz \036z?8\036iG??x?_?9\b\ti?l-K\206P\022hBR(\001\212`\233\226\220\022\004\202a\215\220\202,\205V??\b\230\020?\004???]su?\003\207?cxA\f?~\"<?\n\030Q?}?(?\003?6f|?\222????}1y?\tx?:<.:\214'\022u????? ?\024?\223?+7?F\2360`\225\032\214?0E?\032|?\0036,\0057\004+_/\024y?\224?i\214?F`4\006\216X\207?L?S?\n\016.\t8\022\203?\002?]"...}, static _S_empty_rep_storage = {0, 0, 0, 0}}) at btree.cc:1152 #5 0x0804cd03 in main (argc=2, argv=0xbfffea44) at quartzcompact.cc:370 The aborted of course looks like this: postlist ...Aborted When I compact a test-database with a very simple set of data, quartzcompact does work, also with -F and -n. Best regards, Arjen
Arjen van der Meijden
2005-Jun-24 09:07 UTC
[Xapian-discuss] Quartzcompact in svn-6307 fails
Hmm, appareantly the database wasn't transferred from the original machine to the development one correctly, I'll test it again and I don't think you'd worry too much about this. Although the segfault should not occur of course. Best regards, Arjen On 24-6-2005 9:49, Arjen van der Meijden wrote:> Hi List, > > I'm trying to test some of Olly's changes to the compaction of quartz, > but appareantly quartzcompact just doesn't work at all, or of course my > installation is bugged. > > I've installed svn-6703 on a gentoo system which is quite up-to-date in > terms of gcc (3.3)/glibc versions. > > When I do quartzcompact on our (with 0.8.4) compacted database without > any parameters it just says 'aborted' right after it started, with the > -n and -F parameters the same. > But with -F and -n combined, it segfaults with this backtrace: > > #0 0xb7cf6003 in memmove () from /lib/libc.so.6 > #1 0xb7ee0cdc in Btree::compact (this=0xbfffe780, p=0x8057fa8 > "\004\026?\026?\026?\026?\026?\237") at btree.h:181 > #2 0xb7ee15f0 in Btree::add_item (this=0xbfffe780, j=0) at btree.cc:832 > #3 0xb7ee1b7f in Btree::add_kt (this=0xbfffe780, found=false) at > btree.cc:998 > #4 0xb7ee222f in Btree::add (this=0xbfffe780, key=@0xbfffe6b0, tag> {static npos = 4294967295, _M_dataplus = {<std::allocator<char>> > = {<No data fields>}, _M_p = 0x9933604 > "0??s?g\031\006???\237?]~k?\225%!?\022?\202[gz > \036z?8\036iG??x?_?9\b\ti?l-K\206P\022hBR(\001\212`\233\226\220\022\004\202a\215\220\202,\205V??\b\230\020?\004???]su?\003\207?cxA\f?~\"<?\n\030Q?}?(?\003?6f|?\222????}1y?\tx?:<.:\214'\022u????? > ?\024?\223?+7?F\2360`\225\032\214?0E?\032|?\0036,\0057\004+_/\024y?\224?i\214?F`4\006\216X\207?L?S?\n\016.\t8\022\203?\002?]"...}, > static _S_empty_rep_storage = {0, 0, 0, 0}}) at btree.cc:1152 > #5 0x0804cd03 in main (argc=2, argv=0xbfffea44) at quartzcompact.cc:370 > > The aborted of course looks like this: > postlist ...Aborted > > When I compact a test-database with a very simple set of data, > quartzcompact does work, also with -F and -n. > > Best regards, > > Arjen > > _______________________________________________ > Xapian-discuss mailing list > Xapian-discuss@lists.xapian.org > http://lists.xapian.org/mailman/listinfo/xapian-discuss >