Details
-
Bug
-
Status: Closed (View Workflow)
-
Critical
-
Resolution: Won't Do
-
1.2.3, 1.5.9
-
None
-
Red Hat Enterprise Linux Server release 6.9 (Santiago)
Linux 2.6.32-696.1.1.el6.x86_64 #1 SMP Tue Mar 21 12:19:18 EDT 2017 x86_64 x86_64 x86_64 GNU/Linux
Description
I'm getting issues lately with my web app:
"ERROR","http-nio-8888-exec-30","05/31/2017","10:15:07",""," The specific sequence of files included or processed is: /home/mysite/public_html/wwwroot/requirements/mura/content/contentGatewayAdobe.cfc, line: 731 "
|
(conn:47726) Packets out of order when reading field packets, expected was EOF stream. Packet contents (hex) = 03
|
Query is:
|
|
|
SELECT
|
|
|
title, releasedate, menuTitle, tcontent.lastupdate,summary, tags,tcontent.filename, type,subType, tcontent.siteid,
|
tcontent.contentid, tcontent.contentHistID, target, targetParams,
|
restricted, restrictgroups, displaystart, displaystop, orderno,sortBy,sortDirection,
|
tcontent.fileid, tcontent.credits, tcontent.remoteSource, tcontent.remoteSourceURL, tcontent.remoteURL,
|
tfiles.fileSize,tfiles.fileExt, audience, keypoints
|
,tcontentstats.rating,tcontentstats.totalVotes,tcontentstats.downVotes,tcontentstats.upVotes
|
,tcontentstats.comments, '' as parentType, null as kids,tcontent.path, tcontent.created, tcontent.nextn,
|
tcontent.majorVersion, tcontent.minorVersion, tcontentstats.lockID, tcontentstats.lockType, tcontent.expires,
|
tfiles.filename as AssocFilename,tcontent.displayInterval,tcontent.display,tcontentfilemetadata.altText as fileAltText
|
|
|
FROM tcontent
|
|
|
Left Join tfiles ON (tcontent.fileID=tfiles.fileID)
|
left Join tco...;
|
at org.mariadb.jdbc.internal.util.ExceptionMapper.get(ExceptionMapper.java:150):150
|
at org.mariadb.jdbc.internal.util.ExceptionMapper.getException(ExceptionMapper.java:101):101
|
at org.mariadb.jdbc.internal.util.ExceptionMapper.throwAndLogException(ExceptionMapper.java:77):77
|
at org.mariadb.jdbc.MariaDbStatement.executeQueryEpilog(MariaDbStatement.java:226):226
|
at org.mariadb.jdbc.MariaDbClientPreparedStatement.executeInternal(MariaDbClientPreparedStatement.java:233):233
|
at org.mariadb.jdbc.MariaDbClientPreparedStatement.execute(MariaDbClientPreparedStatement.java:163):163
|
at com.intergral.fusionreactor.jdbc.jdbc42.PreparedStatementSurrogate2.execute(PreparedStatementSurrogate2.java:200):200
|
at lucee.runtime.db.driver.state.StateUtil.execute(StateUtil.java:145):145
|
at lucee.runtime.db.driver.state.StatePreparedStatement.execute(StatePreparedStatement.java:135):135
|
at lucee.runtime.type.util.QueryUtil.execute(QueryUtil.java:240):240
|
at lucee.runtime.type.QueryImpl.<init>(QueryImpl.java:252):252
|
at lucee.runtime.tag.Query.executeDatasoure(Query.java:815):815
|
at lucee.runtime.tag.Query.doEndTag(Query.java:610):610
|
at content.contentgatewayadobe_cfc$cf.udfCall1(/mura/content/contentGatewayAdobe.cfc:731):731
|
at content.contentgatewayadobe_cfc$cf.udfCall(/mura/content/contentGatewayAdobe.cfc):-1
|
These explicit, fatal errors are one of the better types of outcomes during these overall incidents. The most disturbing type of problem that I get is when I don't get an error, but I get crossed queries: Say my app queries, "select a, b, c from mytable", but it ends up with a completely unrelated result (e.g., with columns x, y, and z from some other table), probably from some concurrent connection.
At a glance, it looked like a networking issue to me, so I engaged my CIT department to do some troubleshooting on the network, and although they were monitoring during one of the incidents, they saw nothing wrong.
There is a section of my application that is slow (due to slow queries), and the problem is almost always precipitated by slow requests to this section; however, we haven't been able to reproduce the issue outside of production.
For what it's worth, I've found that the minimum steps to recover from the problem are to:
- Restart the MariaDB service
- Re-initialize the web app (seems to clear out the issues resulting from crossed queries)
This problem is beyond my troubleshooting abilities, since it seems to be low-level (i.e., not related to my application code), but rather related to MariaDB (5.5.56), the JDBC driver (1.2.3 & 1.5.9 – I upgraded to try to fix it, but no luck), or my app server (Lucee--an abstraction layer on top of Java). For what it's worth I also filed a ticket with Lucee, but it's languishing: https://luceeserver.atlassian.net/browse/LDEV-1316
|
data source definition, fwiw |
<data-source allow="127" blob="false" class="org.mariadb.jdbc.Driver" clob="false" connectionLimit="40" |
connectionTimeout="20" custom="" database="" dbdriver="Other" |
dsn="jdbc:mysql://myhost:3306/my_schema?connectTimeout=60000&socketTimeout=180000" host="" |
metaCacheTimeout="60000" name="my_datasource" password="encrypted:mypass" port="-1" storage="false" |
username="my_user" validate="false"/> |
Attachments
Issue Links
- relates to
-
-
- Closed
-