View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0001552 | GNUnet | util library | public | 2010-05-07 11:22 | 2010-07-02 12:37 |
| Reporter | Christian Grothoff | Assigned To | |||
| Priority | high | Severity | major | Reproducibility | random |
| Status | closed | Resolution | unable to reproduce | ||
| Summary | 0001552: socketcall.sendto(msg) points to uninitialized bytes(GNUNET_NETWORK_socket_send) | ||||
| Description | We (sometimes) get a really strange valgrind error (see below). What is particularly strange is that the buffer that valgrind reports to be uninitialized is allocated with GNUnet's malloc, so it should be all zero'ed out at least. Moreover, it should of course actually contain a valid message. | ||||
| Additional Information | ==11384== Syscall param socketcall.sendto(msg) points to uninitialised byte(s) ==11384== at 0x573B6E2: send (send.c:28) ==11384== by 0x5251DB2: GNUNET_NETWORK_socket_send (network.c:474) ==11384== by 0x5241393: transmit_ready (connection.c:1422) ==11384== by 0x52582CF: run_ready (scheduler.c:514) ==11384== by 0x52586DF: GNUNET_SCHEDULER_run (scheduler.c:641) ==11384== by 0x525ED0F: GNUNET_SERVICE_run (service.c:1516) ==11384== by 0x409E48: main (gnunet-service-transport.c:4920) ==11384== Address 0x6f5ce34 is 612 bytes inside a block of size 65,760 alloc'd ==11384== at 0x4C221A7: malloc (vg_replace_malloc.c:195) ==11384== by 0x523BB4E: GNUNET_xmalloc_unchecked_ (common_allocation.c:92) ==11384== by 0x523BA8C: GNUNET_xmalloc_ (common_allocation.c:61) ==11384== by 0x523EF5B: GNUNET_CONNECTION_create_from_accept (connection.c:411) ==11384== by 0x5259142: process_listen_socket (server.c:324) ==11384== by 0x52582CF: run_ready (scheduler.c:514) ==11384== by 0x52586DF: GNUNET_SCHEDULER_run (scheduler.c:641) ==11384== by 0x525ED0F: GNUNET_SERVICE_run (service.c:1516) ==11384== by 0x409E48: main (gnunet-service-transport.c:4920) | ||||
| Tags | No tags attached. | ||||
| related to | 0001554 | closed | core_api sometimes gets strange message instead of INIT_REPLY | |
| related to | 0001553 | closed | Christian Grothoff | datastore testcase crashes with 'content not found' |
|
|
This may or may not be related to transport. If you observe this, please report with the full trace so we can narrow it down (so far only seen with gnunet-service-transport). |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2010-05-07 11:22 | Christian Grothoff | New Issue | |
| 2010-05-07 11:23 | Christian Grothoff | Note Added: 0004014 | |
| 2010-05-07 11:37 | Christian Grothoff | Relationship added | related to 0001554 |
| 2010-05-07 11:38 | Christian Grothoff | Relationship added | related to 0001553 |
| 2010-07-02 12:37 | Christian Grothoff | Status | new => closed |
| 2010-07-02 12:37 | Christian Grothoff | Resolution | open => unable to reproduce |
