[SERVER-236] osx doesn't shutdown cleanly ( WAS: Mongo crashed on OS X - Bad file descriptor ) Created: 14/Aug/09 Updated: 12/Jul/16 Resolved: 29/Aug/09 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Stability |
| Affects Version/s: | None |
| Fix Version/s: | 1.1.0 |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Ryan Nitz | Assignee: | Eliot Horowitz (Inactive) |
| Resolution: | Done | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Environment: |
os x |
||
| Participants: |
| Description |
|
mongodb-osx-x86_64-2009-08-06 Fri Aug 14 14:19:28 MessagingPort recv() error "Bad file descriptor" (9) Fri Aug 14 14:19:28 192.168.2.20r:55765MessagingPort recv() error " Bad file descriptor192.168.2.20:43242192.168.16.152:42966) 9Bad file descriptorBad file descriptorBad file descriptorBad file descriptorFri Aug 14 14:19:28 Fri Aug 14 14:19:28 " ( 192.168.2.20:59878) " (" (" (" (end connection end connection 9Fri Aug 14 14:19:28 Fri Aug 14 14:19:28 192.168.2.20:15570192.168.2.20:43242192.168.16.152:42966end connection Fri Aug 14 14:19:28 192.168.2.20:2008192.168.2.20:17874192.168.2.20:51922192.168.2.20:17106 192.168.2.20:59878end connection Fri Aug 14 14:19:28 Fri Aug 14 14:19:28 closeAllFiles() finished dbexit: really exiting now |
| Comments |
| Comment by Eliot Horowitz (Inactive) [ 15/Sep/09 ] |
|
released in 1.1.0 |
| Comment by Eliot Horowitz (Inactive) [ 29/Aug/09 ] |
|
pretty sure it was a sm gc issue that we fixed |
| Comment by Ryan Nitz [ 26/Aug/09 ] |
|
Upgrading... |
| Comment by Eliot Horowitz (Inactive) [ 26/Aug/09 ] |
|
what version are you running now? |
| Comment by Ryan Nitz [ 26/Aug/09 ] |
|
This just happened again, the difference this time was that a java process/driver disconnected from Mongo. Looks like the process that disconnected was in the middle of an operation. ------------------------------------- Exception in the client: can't say something ------------------------------------- Server: Wed Aug 26 17:47:36 end connection 192.168.2.11:28380 ", inLock: 1, client: "192.168.2.11:51461" } end connection Wed Aug 26 17:54:41 Wed Aug 26 17:54:41 Wed Aug 26 17:54:41 Wed Aug 26 17:54:41 Wed Aug 26 17:54:41 192.168.2.11:41949end connection end connection end connection end connection end connection Wed Aug 26 17:54:41 closeAllFiles() finished ", inLock: 1, client: "192.168.2.11:51461" } |
| Comment by Eliot Horowitz (Inactive) [ 20/Aug/09 ] |
|
ah - i see. |
| Comment by Ryan Nitz [ 20/Aug/09 ] |
|
|
| Comment by Eliot Horowitz (Inactive) [ 20/Aug/09 ] |
|
Did you try to shut down the db and that happened? Or just randomly when a client dies? |
| Comment by Ryan Nitz [ 20/Aug/09 ] |
|
This should be opened again. I can't recreate consistently (just happens). Sync issue? Thu Aug 20 14:42:41 MessagingPort recv() error "Bad file descriptor" (9) 127.0.0.1:60101 ", inLock: 1, client: "192.168.16.152:54392" } |
| Comment by Eliot Horowitz (Inactive) [ 18/Aug/09 ] |
|
closing for now |
| Comment by Ryan Nitz [ 18/Aug/09 ] |
|
Not able to recreate this. My only guess is that I had not properly updated the symlink. I would close it. |
| Comment by Ryan Nitz [ 17/Aug/09 ] |
|
Nothing yet... I will move to the nightly build in the next day or so. |
| Comment by Eliot Horowitz (Inactive) [ 16/Aug/09 ] |
|
any luck? |
| Comment by Ryan Nitz [ 16/Aug/09 ] |
|
Yes, the last stack trace is with 0.9.8 (previous ones with older nightly build). Let me see if I can consistently reproduce.... |
| Comment by Eliot Horowitz (Inactive) [ 15/Aug/09 ] |
|
That's definitely with 0.9.8? |
| Comment by Ryan Nitz [ 15/Aug/09 ] |
|
Nope... still a problem. The problem seems to be when the client disconnects and then reconnects right away. Sat Aug 15 11:46:33 Got signal: 11 (Segmentation fault). Sat Aug 15 11:46:33 Backtrace: |
| Comment by Ryan Nitz [ 15/Aug/09 ] |
|
K... it ran without failure. I don't have permission to assign/close bugs so you will have to handle. |
| Comment by Ryan Nitz [ 15/Aug/09 ] |
|
I upgraded last night. I'll run the code that was causing the failure and let you know. |
| Comment by Eliot Horowitz (Inactive) [ 15/Aug/09 ] |
|
any update on this? |
| Comment by Eliot Horowitz (Inactive) [ 14/Aug/09 ] |
|
fairly sure this was fixed in 0.9.8 |
| Comment by Ryan Nitz [ 14/Aug/09 ] |
|
More info... switching to latest release to see if it persists. Fri Aug 14 18:02:06 CMD: drop lindex.sites ", inLock: 1, client: "127.0.0.1:61175" } |