Problem #3 from bug 4347. Currently, if you are running in SSL mode, you will get a warning after logout/disconnect: ERROR: SSL_read: 5 (Förbindelse borttagen av partnern) ERROR: SSL_write: 5 (Brutet rör) This is harmless, since the return code is zero. It is however somewhat ugly and confusing, so this should be silenced.
This issue does only occur on Win2003 Terminal Server in the following codepath of rdesktop: (gdb) r Starting program: /home/hean01/Development/upstream/rdesktop.git/rdesktop 10.47.254.170 -u administrator -p Vogh7Grot [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib64/libthread_db.so.1". Autoselected keyboard map en-us Connection established using SSL. WARNING: Remote desktop does not support colour depth 24; falling back to 16 bt ERROR: SSL_read: 5 (Connection reset by peer) Program received signal SIGPIPE, Broken pipe. 0x000000334500e030 in __write_nocancel () at ../sysdeps/unix/syscall-template.S:82 82 T_PSEUDO (SYSCALL_SYMBOL, SYSCALL_NAME, SYSCALL_NARGS) Missing separate debuginfos, use: debuginfo-install libXcursor-1.1.13-1.fc17.x86_64 libXfixes-5.0-2.fc17.x86_64 pcsc-lite-libs-1.8.3-1.fc17.x86_64 (gdb) bt #0 0x000000334500e030 in __write_nocancel () at ../sysdeps/unix/syscall-template.S:82 #1 0x000000334dcaff15 in sock_write (b=0xa253f0, in=0xa2f723 "\027\003\001", inl=32) at bss_sock.c:158 #2 0x000000334dcadcf9 in BIO_write (b=0xa253f0, in=0xa2f723, inl=32) at bio_lib.c:247 #3 0x000000383aa24fc2 in ssl3_write_pending (s=s@entry=0xa24ff0, type=type@entry=23, buf=<optimized out>, len=<optimized out>) at s3_pkt.c:837 #4 0x000000383aa25444 in do_ssl3_write (s=s@entry=0xa24ff0, type=type@entry=23, buf=buf@entry=0xa05750 "\003", len=<optimized out>, create_empty_fragment=create_empty_fragment@entry=0) at s3_pkt.c:809 #5 0x000000383aa255b3 in ssl3_write_bytes (s=0xa24ff0, type=23, buf_=0xa05750, len=<optimized out>) at s3_pkt.c:604 #6 0x000000000041ca54 in tcp_send (s=0x6c7970) at tcp.c:130 #7 0x000000000041d5da in iso_send_msg (code=code@entry=128 '\200') at iso.c:48 #8 0x000000000041db6e in iso_disconnect () at iso.c:346 #9 0x000000000041e225 in mcs_disconnect () at mcs.c:358 #10 0x000000000041f705 in sec_disconnect () at secure.c:913 #11 0x0000000000422015 in rdp_disconnect () at rdp.c:1666 #12 0x0000000000407a5e in main (argc=<optimized out>, argv=<optimized out>) at rdesktop.c:1026
Verified that this is a problem when using plain RDP too, there is no proper handling of disconnects and when rdp loop exists it will try send data on a closed socket. I could produce this case by connect using plain rdp and then disconnect network cabel from server while producing rdp traffic.
Fixed in upstream commit r1681
Vendordrop r26147 brings the changes for this bug. Testcase is descibed in comment #2.
The test case in comment #2 results in: [tluser@dhcp-254-223 ~]$ tl-run-rdesktop Connecting to RDP server WIN-FGI326MKOI9... PRINTER PRN1 to nearest driver HP Color LaserJet 8500 PS PRINTER PRN2 to thinlocal driver HP Color LaserJet 8500 PS Autoselected keyboard map en-gb Failed to negotiate protocol, retrying with plain RDP. WARNING: Remote desktop does not support colour depth 24; falling back to 16 ERROR: send: No route to host ERROR: send: Broken pipe ERROR: send: Broken pipe ERROR: send: Broken pipe ERROR: send: Broken pipe ERROR: send: Broken pipe ERROR: send: Broken pipe ERROR: send: Broken pipe ERROR: send: Broken pipe ERROR: send: Broken pipe ERROR: send: Broken pipe And eventually: ERROR: send: Broken pipe ERROR: send: Broken pipe ERROR: Connection closed ERROR: connect: No route to host Failed to connect, retry in 4 secs... ERROR: connect: No route to host Failed to connect, retry in 4 secs... ERROR: connect: No route to host Failed to connect, retry in 4 secs... ERROR: connect: No route to host Failed to connect, retry in 4 secs... ERROR: connect: No route to host Failed to connect, retry in 4 secs...
The original problem doesn't seem to happen though. Disconnecting a TLS session does not result in any extra errors.
(In reply to comment #5) > The test case in comment #2 results in: > > [tluser@dhcp-254-223 ~]$ tl-run-rdesktop > Connecting to RDP server WIN-FGI326MKOI9... > PRINTER PRN1 to nearest driver HP Color LaserJet 8500 PS > PRINTER PRN2 to thinlocal driver HP Color LaserJet 8500 PS > Autoselected keyboard map en-gb > Failed to negotiate protocol, retrying with plain RDP. > WARNING: Remote desktop does not support colour depth 24; falling back to 16 > ERROR: send: No route to host > ERROR: send: Broken pipe > ERROR: send: Broken pipe > ERROR: send: Broken pipe > ERROR: send: Broken pipe > ERROR: send: Broken pipe > ERROR: send: Broken pipe > ERROR: send: Broken pipe > ERROR: send: Broken pipe > ERROR: send: Broken pipe > ERROR: send: Broken pipe > > And eventually: > > ERROR: send: Broken pipe > ERROR: send: Broken pipe > ERROR: Connection closed > ERROR: connect: No route to host > Failed to connect, retry in 4 secs... > ERROR: connect: No route to host > Failed to connect, retry in 4 secs... > ERROR: connect: No route to host > Failed to connect, retry in 4 secs... > ERROR: connect: No route to host > Failed to connect, retry in 4 secs... > ERROR: connect: No route to host > Failed to connect, retry in 4 secs... Fixed upstream in r1711, will vendor drop soon.
(In reply to comment #7) > (In reply to comment #5) > > The test case in comment #2 results in: > > > > [tluser@dhcp-254-223 ~]$ tl-run-rdesktop > > Connecting to RDP server WIN-FGI326MKOI9... > > PRINTER PRN1 to nearest driver HP Color LaserJet 8500 PS > > PRINTER PRN2 to thinlocal driver HP Color LaserJet 8500 PS > > Autoselected keyboard map en-gb > > Failed to negotiate protocol, retrying with plain RDP. > > WARNING: Remote desktop does not support colour depth 24; falling back to 16 > > ERROR: send: No route to host > > ERROR: send: Broken pipe > > ERROR: send: Broken pipe > > ERROR: send: Broken pipe > > ERROR: send: Broken pipe > > ERROR: send: Broken pipe > > ERROR: send: Broken pipe > > ERROR: send: Broken pipe > > ERROR: send: Broken pipe > > ERROR: send: Broken pipe > > ERROR: send: Broken pipe > > > > And eventually: > > > > ERROR: send: Broken pipe > > ERROR: send: Broken pipe > > ERROR: Connection closed > > ERROR: connect: No route to host > > Failed to connect, retry in 4 secs... > > ERROR: connect: No route to host > > Failed to connect, retry in 4 secs... > > ERROR: connect: No route to host > > Failed to connect, retry in 4 secs... > > ERROR: connect: No route to host > > Failed to connect, retry in 4 secs... > > ERROR: connect: No route to host > > Failed to connect, retry in 4 secs... > > Fixed upstream in r1711, will vendor drop soon. Vendordrop in commit r27523.
Still getting a whole bunch of SSL errors in the timeout case. This time it seems like they're from OpenSSL though, so we might have to live with them: [tluser@dhcp-254-223 Desktop]$ /opt/thinlinc/bin/rdesktop WIN-FGI326MKOI9 Autoselected keyboard map en-gb warning: unable to open /etc/gssapi_mech.conf: errno 2 (No such file or directory) WARNING: CredSSP: System doesn't have support for desired authentication mechanism. Connection established using SSL. WARNING: Remote desktop does not support colour depth 24; falling back to 16 ERROR: SSL_write: 5 (No route to host) 140359017301696:error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry:s3_pkt.c:866: 140359017301696:error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry:s3_pkt.c:866: 140359017301696:error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry:s3_pkt.c:866: 140359017301696:error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry:s3_pkt.c:866: 140359017301696:error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry:s3_pkt.c:866: 140359017301696:error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry:s3_pkt.c:866: 140359017301696:error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry:s3_pkt.c:866: 140359017301696:error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry:s3_pkt.c:866: 140359017301696:error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry:s3_pkt.c:866: 140359017301696:error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry:s3_pkt.c:866: 140359017301696:error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry:s3_pkt.c:866: 140359017301696:error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry:s3_pkt.c:866: 140359017301696:error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry:s3_pkt.c:866: 140359017301696:error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry:s3_pkt.c:866: 140359017301696:error:1409F07F:SSL routines:SSL3_WRITE_PENDING:bad write retry:s3_pkt.c:866: Disconnected due to network error, retrying to reconnect for 70 minutes. ERROR: send: Connection reset by peer Opening for discussion.
Needs retesting
Seems to be fixed in the new code.
