[Date Prev][Date Next]
Re: Bugfix: wait4msg() hangs when using SSL/TLS (ITS#446)
Forwarded to devel for discussion....
At 02:35 AM 3/20/00 +0000, Andrew Hacking wrote:
>I guess this means the socket is in 'blocking' mode, which would always cause a
>hang when reading a byte that is not yet available (as I think you have shown).
Yes, but why did it attempt to read a PDU from the slave? It appears to me
that your change caused the library to read from the wrong connection.
>This is quite serious, since even if a select() is performed to see if _some_ data
>is available BEFORE calling try_read1msg(), the stream_read() will hang if not
>enough bytes have been queued on the socket when attempting to parse the message.
>I always thought/assumed (I know, assumption is never a good thing) that sockbuf's
>used non-blocking sockets for this very reason... to avoid hanging indefinitely
>when performing a read/recv.
No. By default the library uses blocking sockets.
>This means that performing a read/recv on a
>non-blocking socket when data is unavailable returns failure with errno set to
>EWOULDBLOCK/EAGAIN, then select() would be called to wait for the arrival of more
>data, the timeout checked, and the read/recv tried again.
Only if the caller marks the sockets non-blocking.
>So the question is, why are blocking sockets being used, or has it always been this
>way and no-one noticed ?
Because UNIX socket model is, by default, blocking.
>I am a little concerned because libldap and liblber appear to be written around the
>notion of non-blocking sockets.
libldap and liblber support both blocking and non-blocking sockets.
>If blocking sockets are used, the timeout that the caller provides to libldap will
>never be honoured when say, the server takes some time to respond, or worse it
>crashes, it will even hang when network congestion/disruption occurs.
The timeout applies only to the select, not to the read. Once a PDU read is
initiated, it must be completed (or the stream closed). The API doesn't
support restarting of PDU reads.
>It is under
>all these conditions that the timeout ensures libldap only blocks for a definite
>amount of time.
>> I've experimented with this patch a bit (without TLS) and found
>> that it causes test009-referral to hang. ldapsearch has data
>> pending from the master but is blocked reading from the slave.
>> (gdb) where
>> #0 0x18183f44 in read () from /usr/lib/libc.so.3
>> #1 0x805a3d5 in stream_read (sb=0x806a000, buf=0x806718c, len=1)
>> at ../../../libraries/liblber/sockbuf.c:911
>> #2 0x8059df3 in ber_pvt_sb_read (sb=0x806a000, buf_arg=0x806718c, len=1)
>> at ../../../libraries/liblber/sockbuf.c:504
>> #3 0x8058b6b in ber_get_next (sb=0x806a000, len=0xbfbfcb70, ber=0x8067180)
>> at ../../../libraries/liblber/io.c:525
>> #4 0x804ca47 in try_read1msg (ld=0x806a000, msgid=-1, all=1, sb=0x806a000,
>> lc=0x8061100, result=0xbfbfcc34)
>> at ../../../libraries/libldap/result.c:296
>> #5 0x804c890 in wait4msg (ld=0x806a000, msgid=-1, all=1, timeout=0x0,
>> result=0xbfbfcc34) at ../../../libraries/libldap/result.c:208
>> #6 0x804c714 in ldap_result (ld=0x806a000, msgid=-1, all=1, timeout=0x0,
>> result=0xbfbfcc34) at ../../../libraries/libldap/result.c:137
>> #7 0x804aef7 in dosearch (ld=0x806a000,
>> base=0x8061040 "o=University of Michigan, c=US", scope=2, attrs=0x0,
>> attrsonly=0, filtpatt=0x0, value=0x8062050 "sn=jensen")
>> at ../../../clients/tools/ldapsearch.c:540
>> #8 0x804acfb in main (argc=11, argv=0xbfbfd4e8)
>> at ../../../clients/tools/ldapsearch.c:471
>> #9 0x804a095 in _start ()
>> At 05:42 AM 2/8/00 GMT, email@example.com wrote:
>> >Full_Name: Andrew Hacking
>> >Version: -devel
>> >OS: Linux
>> >URL: ftp://ftp.openldap.org/incoming/andrew-hacking-000208-2.patch
>> >Submission from: (NULL) (126.96.36.199)
>> >Sorry for the re-post, my email address was changed today...
>> >libldap wait4msg() hangs under various conditions when using SSL/TLS.
>> >The specifics are that wait4msg blocks in do_ldap_select()
>> >while waiting for data that has _already_ been buffered by the
>> >OpenSSL library. do_ldap_select() should never be called until
>> >the SSL/TLS buffers have been exhausted.
>> >The patch changes to wait4msg() changes the way messages are read to ensure the
>> >SSL/TLS buffers really are exhausted before calling do_ldap_select()
>> >try_read1msg() is also changed to return 'exhausted' (-3) on an
>> >EWOULDBLOCK/EAGAIN condition. This differentiates EWOULDBLOCK/EAGAIN
>> >(ie the underlying sockbuf is now exhausted) with the 'keep looking' (-2)