| Summary: | ASTERISK-28649: Segfault: ast_variables_destroy #channel_vars #set_var #sorcery_realtime | ||||
| Reporter: | John Bowen (jbowen7@gmail.com) | Labels: | |||
| Date Opened: | 2019-12-04 14:22:31.000-0600 | Date Closed: | 2019-12-05 13:13:05.000-0600 | ||
| Priority: | Major | Regression? | Yes | ||
| Status: | Closed/Complete | Components: | Resources/res_realtime | ||
| Versions: | 16.5.0 16.6.2 | Frequency of Occurrence | Constant | ||
| Related Issues: |
| ||||
| Environment: | CentOS Linux release 7.6.1810 (Core) 3.10.0-957.27.2.el7.x86_64 Asterisk build: asterisk-16.6.2 Alembic Version: 3a094a18e75b | Attachments: | ( 0) core.28564-brief.txt ( 1) core.28564-full.txt ( 2) core.28564-locks.txt ( 3) core.28564-thread1.txt | ||
| Description: | Using sorcery realtime, if the column ps_endpoints.set_var` contains more than 16 characters a segfault occurs during the loading of endpoints.
This, however, does not occur when statically defining endpoints in pjsip.conf. Reproduce with: 1) Create three endpoints (odbc, realtime) 2) UPDATE ps_endpoints SET set_var = 'abc=1234567890abc' WHERE id is NOT NULL 3) Reload asterisk Notes: - All 3 endpoints must have channel_vars > 16 characters, the issue is not reproducable if only 2 endpoints have channel_vars > 16 chars - This issue was not present in 13.17.2 (using pjsip bundled) I'll be posting my findings as I dig into the code, but wanted to make core and other users aware of the issue first. | ||||
| Comments: | By: Asterisk Team (asteriskteam) 2019-12-04 14:22:32.600-0600 Thanks for creating a report! The issue has entered the triage process. That means the issue will wait in this status until a Bug Marshal has an opportunity to review the issue. Once the issue has been reviewed you will receive comments regarding the next steps towards resolution. A good first step is for you to review the [Asterisk Issue Guidelines|https://wiki.asterisk.org/wiki/display/AST/Asterisk+Issue+Guidelines] if you haven't already. The guidelines detail what is expected from an Asterisk issue report. Then, if you are submitting a patch, please review the [Patch Contribution Process|https://wiki.asterisk.org/wiki/display/AST/Patch+Contribution+Process]. Please note that once your issue enters an open state it has been accepted. As Asterisk is an open source project there is no guarantee or timeframe on when your issue will be looked into. If you need expedient resolution you will need to find and pay a suitable developer. Asking for an update on your issue will not yield any progress on it and will not result in a response. All updates are posted to the issue when they occur. By: John Bowen (jbowen7@gmail.com) 2019-12-04 18:34:49.700-0600 This might be a problem in ODBC. (if it is, i'll close the issue soon) When I alter the column type from TEXT -> VARCHAR(255) the segfault no longer occurs. LONGTEXT and TEXT both produce the segfault, whereas VARCHAR does not. Disregard "This issue was not present in 13.17.2 (using pjsip bundled)", since I was using different ODBC drivers. I'll report more findings shortly Installed: Name : unixODBC Arch : x86_64 Version : 2.3.1 Release : 11.el7 From repo : base Name : unixODBC-devel Arch : x86_64 Version : 2.3.1 Release : 11.el7 From repo : base Name : mysql-connector-odbc Arch : x86_64 Version : 8.0.18 Release : 1.el7 From repo : mysql-connectors-community By: John Bowen (jbowen7@gmail.com) 2019-12-04 22:57:01.830-0600 More info: When logging the values in "pjsip_configuration.set_var_handler" I see the values appear to be identical with both column types: TEXT and VARCHAR. However, when I switch to VARCHAR, a call to set_var_to_str occurs, whereas with TEXT it does not. By: Sean Bright (seanbright) 2019-12-05 08:17:16.331-0600 Pretty sure this is the same bug as the linked issue ASTERISK-28112. Per [the MySQL bug linked from the other issue|https://bugs.mysql.com/bug.php?id=92947], setting {{NO_SSPS = 1}} in your {{odbc.ini}} file might resolve the issue. It did in my tests. By: John Bowen (jbowen7@gmail.com) 2019-12-05 13:07:49.267-0600 @Sean, my preliminary tests show that adding `NO_SSPS =1` does indeed resolve the issue. Can't say I know why yet, but all fingers are pointing at the driver. I'm going to go ahead and close this | ||||