ES 2.2 Just installed, might have problems

sixlxvi

Well-known member
Just installed XFES after installing Elasticsearch on the server.

A few seconds after installing, my index page and admin CP displayed "an unknown database error occurred" but then it fixed itself a few seconds later. But then it happened again.

Also, I when the site loaded, I made the minimum search amount 2 characters. I did a few search tests but they come back with no results.

Edit: Just got this as well:
Time: Wed May 11 16:31:44 2022 -0400
Account: elasticsearch
Resource: Process Time
Exceeded: 1856 > 1800 (seconds)
Executable: /usr/share/elasticsearch/modules/x-pack-ml/platform/linux-x86_64/bin/controller
Command Line: /usr/share/elasticsearch/modules/x-pack-ml/platform/linux-x86_64/bin/controller
PID: 351458 (Parent PID:351154)
Killed: No

Edit 2: When I test settings in the admin panel, I get "A connection could not be made to the Elasticsearch server."

Talking to server admin right now.
 

Brogan

XenForo moderator
Staff member
A few seconds after installing, my index page and admin CP displayed "an unknown database error occurred"
If it happens again, view the page source and there should be more details.

I made the minimum search amount 2 characters. I did a few search tests but they come back with no results.
Did you rebuild the search index after enabling XFES in the ACP?

Just got this as well:
Looks like some server set up and optimisation may be required.
 

Brogan

XenForo moderator
Staff member
If using XFES you can truncate the table as it's not used.

 

sixlxvi

Well-known member
If using XFES you can truncate the table as it's not used.

Cool, I didn't realize that was another perk of XFES. Do I still need to rebuild before/after truncating or does the rebuild become obsolete anyway?

Speaking of rebuilding, I also upgraded to PHP 8.1 today, and switched from MySQL 5.7 to MariaDB 10.3 - should I rebuild anything in XF because of that?
 

Brogan

XenForo moderator
Staff member
Yes, you will need to rebuild as ES currently doesn't have any content indexed.

Speaking of rebuilding, I also upgraded to PHP 8.1 today, and switched from MySQL 5.7 to MariaDB 10.3 - should I rebuild anything in XF because of that?
No rebuilding is required.

If you use third party add-ons, confirm they are compatible.
 

sixlxvi

Well-known member
What are you searching for?

Do you have any stop words set?

View attachment 268147
No stop words set.

I searched for "oil" (it's an auto forum) and "t3g" which is the acronym for the site.

But I just got another email from the server, maybe that is why?:
Time: Wed May 11 16:54:45 2022 -0400
Account: elasticsearch
Resource: RSS Memory Size
Exceeded: 3378 > 500 (MB)
Executable: /usr/share/elasticsearch/jdk/bin/java
Command Line: /usr/share/elasticsearch/jdk/bin/java -Xshare:auto -Des.networkaddress.cache.ttl=60 -Des.networkaddress.cache.negative.ttl=10 -Djava.security.manager=allow -XX:+AlwaysPreTouch -Xss1m -Djava.awt.headless=true -Dfile.encoding=UTF-8 -Djna.nosys=true -XX:-OmitStackTraceInFastThrow -XX:+ShowCodeDetailsInExceptionMessages -Dio.netty.noUnsafe=true -Dio.netty.noKeySetOptimization=true -Dio.netty.recycler.maxCapacityPerThread=0 -Dlog4j.shutdownHookEnabled=false -Dlog4j2.disable.jmx=true -Dlog4j2.formatMsgNoLookups=true -Djava.locale.providers=SPI,COMPAT --add-opens=java.base/java.io=ALL-UNNAMED -XX:+UseG1GC -Djava.io.tmpdir=/tmp/elasticsearch-14648368063889456075 -XX:+HeapDumpOnOutOfMemoryError -XX:+ExitOnOutOfMemoryError -XX:HeapDumpPath=/var/lib/elasticsearch -XX:ErrorFile=/var/log/elasticsearch/hs_err_pid%p.log -Xlog:gc*,gc+age=trace,safepoint:file=/var/log/elasticsearch/gc.log:utctime,pid,tags:filecount=32,filesize=64m -Xms3072m -Xmx3072m
-XX:MaxDirectMemorySize=1610612736 -XX:G1HeapRegionSize=4m -XX:InitiatingHeapOccupancyPercent=30 -XX:G1ReservePercent=15 -Des.path.home=/usr/share/elasticsearch -Des.path.conf=/etc/elasticsearch -Des.distribution.flavor=default -Des.distribution.type=rpm -Des.bundled_jdk=true -cp /usr/share/elasticsearch/lib/* org.elasticsearch.bootstrap.Elasticsearch -p /var/run/elasticsearch/elasticsearch.pid
--quietPID: 874379 (Parent PID:874379)
Killed: No

Still waiting on my server provider to reply about that.
 

sixlxvi

Well-known member
Does this seem accurate to you?

I am really sorry. Unfortunately, we hit a wall. The vm.max_map_count value needed for elasticsearch to start, is not changeable in an OpenVZ server. I had to find this issue looking for a solution


I would recommend in case you will be running elasticsearch to migrate the server to a Cloud Server which gives you the opportunity to have full control of the server kernel.
 

sixlxvi

Well-known member
So nothing is wrong with XenForo.

The problem is that ElastSearch needs to modify some kernel parameters to run OK, which is not possible on an OpenVZ server. It is totally fine on a Cloud server by the way.

In order to change the Kernel values for your VPS, we will need to change them in the container, which would affect the whole server fleet on that container. I think you won't lose anything going to a Cloud server. You will have the possibility to install not only elasticsearch service but any other service that needs to modify the kernel settings.
 

sixlxvi

Well-known member
For future readers of this thread:

Issues were solved server side. It was a problem with the Elasticsearch install.

@Brogan - Is the option "Empty the MySQL search index" an automatic version of the table truncate?
 
Top