I'm experiencing a suspiciously similar segfault on 2.4.44+dfsg-1
Here follows the backtrace:
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7f60dd7e1700 (LWP 32510)]
0x00007f60dd7e0aa0 in ?? ()
(gdb) bt
#0 0x00007f60dd7e0aa0 in ?? ()
#1 0x00007f61650bca41 in slap=5Fwritewait=5Fplay (op=3D0x7f60c8002550)=
at ../../../../servers/slapd/result.c:294
#2 send=5Fldap=5Fber (op=3D0x7f60c8002550, ber=3D0x7f60dd64f250) at ..=
/../../../servers/slapd/result.c:367
#3 0x00007f61650bf651 in slap=5Fsend=5Fsearch=5Fentry (op=3D0x7f60c800=
2550, rs=3D0x7f60dd7e0aa0) at ../../../../servers/slapd/result.c:1430
#4 0x00007f616003590b in mdb=5Fsearch (op=3D0x7f60c8002550, rs=3D0x7f6=
0dd7e0aa0) at ../../../../../servers/slapd/back-mdb/search.c:1086
#5 0x00007f615f5f8cd6 in relay=5Fback=5Fop (op=3D0x7f60c8002550, rs=3D=
0x7f60dd7e0aa0, which=3D<optimized out>)
at ../../../../../servers/slapd/back-relay/op.c:210
#6 0x00007f616511aeea in overlay=5Fop=5Fwalk (op=3Dop@entry=3D0x7f60c8=
002550, rs=3D0x7f60dd7e0aa0, which=3Dop=5Fsearch, oi=3D0x7f6165d2c020,=
on=3D<optimized out>) at ../../../../servers/slapd/backover.c:677
#7 0x00007f616511b044 in over=5Fop=5Ffunc (op=3D0x7f60c8002550, rs=3D<=
optimized out>, which=3D<optimized out>)
at ../../../../servers/slapd/backover.c:730
#8 0x00007f61650af071 in fe=5Fop=5Fsearch (op=3D0x7f60c8002550, rs=3D0=
x7f60dd7e0aa0) at ../../../../servers/slapd/search.c:402
#9 0x00007f61650ae9ee in do=5Fsearch (op=3D0x7f60c8002550, rs=3D0x7f60=
dd7e0aa0) at ../../../../servers/slapd/search.c:247
#10 0x00007f61650ac57c in connection=5Foperation (ctx=3D0x7f60dd7e0c10,=
arg=5Fv=3D0x7f60c8002550)
at ../../../../servers/slapd/connection.c:1158
#11 0x00007f61650ac867 in connection=5Fread=5Fthread (ctx=3D0x7f60c8002=
550, argv=3D0x7f60dd7df710)
at ../../../../servers/slapd/connection.c:1294
#12 0x00007f6164c0df22 in ldap=5Fint=5Fthread=5Fpool=5Fwrapper (xpool=3D=
0x7f6165c8ffa0) at ../../../../libraries/libldap=5Fr/tpool.c:696
#13 0x00007f61631f20a4 in start=5Fthread (arg=3D0x7f60dd7e1700) at pthr=
ead=5Fcreate.c:309
#14 0x00007f6162f2762d in clone () at ../sysdeps/unix/sysv/linux/x86=5F=
64/clone.S:111
To reliably trigger the segfault I need to request jpegImage attributes=
. A few more (possibly important) facts:
the server houses two databases, the first is a sync-repl slave unsing =
the mdb backend, the second a relay backend
that provides read-only access to the first database rewriting the data=
base suffix.
So far I could only trigger the segfault in the relay database, not in =
the mdb-backend one.
HTH Ralf Mattes