[ZBX-5565] API returns HTML page when the database is down, instead of returning an appropriate error response. Created: 2012 Sep 10 Updated: 2017 May 30 Resolved: 2012 Oct 22 |
|
Status: | Closed |
Project: | ZABBIX BUGS AND ISSUES |
Component/s: | API (A) |
Affects Version/s: | 2.0.2 |
Fix Version/s: | 2.1.0 |
Type: | Incident report | Priority: | Major |
Reporter: | Kristoffer Berdal | Assignee: | Unassigned |
Resolution: | Fixed | Votes: | 0 |
Labels: | api, json, usability | ||
Remaining Estimate: | Not Specified | ||
Time Spent: | Not Specified | ||
Original Estimate: | Not Specified | ||
Environment: |
Does it matter? Centos |
Description |
The webpage that is shown to the user when zabbix is unable to connect to the database is also returned when you make a call to the API while the database is down. This causes problems as instead of getting JSON you are suddenly getting HTML to parse. This is the text on the website: pg_connect() [<a href='function.pg-connect'>function.pg-connect</a>]: Unable to connect to PostgreSQL server: could not connect to server: Connection refused Zabbix is temporarily unavailable! This is not good. The API should return an error that can easily be parsed, just like the rest of them. Preferrably it would be best if the API started defining its own error codes as the JSONRPC v2 standard says you should. http://www.jsonrpc.org/specification#error_object I do not know if this is fixed in a later version, but I noticed the problem when I suddenly got an error in the simple API client I have created for node.js https://github.com/flexd/zabbix.js Getting HTML when you expect to parse JSON and no errors were returned is not good |
Comments |
Comment by Oleksii Zagorskyi [ 2012 Sep 24 ] |
the "warning.php:29" error also mentioned in |
Comment by Kristoffer Berdal [ 2012 Sep 24 ] |
That is unrelated to the problem here. The problem is that it should not be showing the error HTML page in the first place! |
Comment by Alexey Fukalov [ 2012 Oct 11 ] |
dev branch: svn://svn.zabbix.com/branches/dev/ZBX-5565 |
Comment by Eduards Samersovs (Inactive) [ 2012 Oct 17 ] |
(1) without zabbix.conf.php Frontend doesn't work Similar fatal error if we open http://localhost/warning.php |
Comment by Eduards Samersovs (Inactive) [ 2012 Oct 17 ] |
(2) Offer to use CWarning class in warning.php Vedmak RESOLVED |
Comment by Eduards Samersovs (Inactive) [ 2012 Oct 19 ] |
Tested! |
Comment by Alexey Fukalov [ 2012 Oct 19 ] |
Fixed in pre-2.1.0 r30948. |
Comment by Pavels Jelisejevs (Inactive) [ 2012 Oct 19 ] |
(3) Please note this in the API changelog under the general section. Vedmak CLOSED |
Comment by Alexander Vladishev [ 2012 Oct 19 ] |
(5) Errors on all pages like: Warning: array_map() [function.array-map]: The first argument, 'stripslashes_deep', should be either NULL or a valid callback in include/defines.inc.php on line 927 Warning: array_map() [function.array-map]: The first argument, 'stripslashes_deep', should be either NULL or a valid callback in include/defines.inc.php on line 927 Warning: array_map() [function.array-map]: The first argument, 'stripslashes_deep', should be either NULL or a valid callback in include/defines.inc.php on line 927 Cannot modify header information - headers already sent by (output started at include/defines.inc.php:927) [func.inc.php:70] Cannot use a scalar value as an array [page_header.php:28] Cannot modify header information - headers already sent by (output started at include/defines.inc.php:927) [page_header.php:105] Vedmak RESOLVED sasha CLOSED |
Comment by Kristoffer Berdal [ 2012 Oct 22 ] |
What's going on with this? I saw it was closed and fixed a long time ago. Why the bloody hell am I getting like 20 messages of email spam per day about this? |
Comment by richlv [ 2012 Oct 22 ] |
the fix is being polished and improved. developers do work on it. as it's all opensource, you get to follow all the development process. |
Comment by Oleksii Zagorskyi [ 2012 Oct 22 ] |
Kristoffer, you are issue reporter, that's why you are receiving notifications. |
Comment by Alexey Fukalov [ 2012 Oct 22 ] |
Fixed in pre-2.1.0 r31014. |
Comment by richlv [ 2013 Jan 08 ] |
this resulted in a regression : |
Comment by richlv [ 2013 Sep 30 ] |
this resulted in another regression : |