AlexJ

@InvisionHQ - Is this theme issue or addon issue? using @Ehren elegant theme:

Are the reported issues resolved? @iacas  @InvisionHQ - Should i hold on upgrade? thanks

Thanks @Jordan Miller for sharing update. Changes look better then before. I agree with @AlexWebsites idea.. and if possible and option with multi-select - so users can search both Pages and Forum together OR forum and downloads, etc. 

9 hours ago, Stuart Silvester said:

The pre-selected option is contextual depending on the page that the user is viewing. This is very similar to how it used to work in 3.x and early versions of 4.x.

 

5 hours ago, Ehren said:

Thanks guys - the width of the search menu has been increased here. I have a few slight improvements I'd like to make which I'll discuss with the rest of the team shortly.

Are we going to loose below current existing nice drop with new one which looks ugly + limited in functionality? 

Just now, Stuart Silvester said:

We're using a custom theme, we need to make a few adjustments.

It looks like this in the default theme

Is it multiselect with Everywhere/All has default? I added 2nd point while you responded so if you can please share an update on selection - that would be great. 

How does the drop down look in default theme?  

Thanks

Noticed update in release notes for Search UI. 

1. Just showing 3-4 letters as a part of search is not a good idea. I am using 2K monitor. Can we please improve UI for Search?

2. Also needs to be multi-select -> User is not going to search one time in pages and then in topics for example. Default can be: ALL but rest all multiselect option.

Thanks. 

Hello,

@SoftwareFactory How to fix this one? You had helped in past to move to custom to load the custom fix. I would like to upgrade to latest version which you have released.

I keep seeing below error's randomly but more often... Any thoughts on why it's happening? 

IPS\Http\Request\CurlException: https://api.giphy.com/v1/gifs/trending?api_key=XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX&limit=30&offset=0&lang=en&rating=pg-13
Connection timed out after 10001 milliseconds (28)
#0 /home/xxxx/public_html/system/Http/Request/Curl.php(422): IPS\Http\Request\_Curl->_execute()
#1 /home/xxxx/public_html/system/Http/Request/Curl.php(298): IPS\Http\Request\_Curl->_executeAndFollowRedirects()
#2 /home/xxxx/public_html/applications/core/modules/front/system/editor.php(419): IPS\Http\Request\_Curl->get()
#3 /home/xxxx/public_html/system/Dispatcher/Controller.php(107): IPS\core\modules\front\system\_editor->giphy()
#4 /home/xxxx/public_html/system/Dispatcher/Dispatcher.php(153): IPS\Dispatcher\_Controller->execute()
#5 /home/xxxx/public_html/index.php(13): IPS\_Dispatcher->run()
#6 {main}

Anh - that 2nd URL makes sense.

I don't have issues with TCP - since for email, I use AWS API. Plus all my TCP connections are mostly available through Cloudflare. To restrict DDoS events, I just had banned multiple data-center ASN plus it helps with spam protection as well, since majority of spammers, use Proxy. 

mysql> SHOW VARIABLES LIKE '%log_warn%';
+---------------+-------+
| Variable_name | Value |
+---------------+-------+
| log_warnings  | 2     |
+---------------+-------+
1 row in set (0.00 sec)

Time to set to 1 and monitor. 

Thanks for the help. 

On 7/17/2022 at 12:19 AM, Randy Calvert said:

By the way… hit Google for that phrase “Got timeout reading communication packets”. It’s not just us saying this. 😉 

I had tried Google approach first and didn't work and hence created topic 😞 

Sadly, didn't help after making it back to 60 sec. So far moved from 90 sec to 15 sec and back to 60 sec. 

Error still remains. Any other tips, to pin point it? Thank you.

I noticed similar issue - check addons.. some addons are not compatible with php 8.1. 

I have no idea why it was removed.. it was just quicker to run query... since I don't need any another bloatware of phpMyadmin... need to stick with shell now. 

It was handy feature. 

1 hour ago, DanielFb said:

Is 4.7 going to be come compatible?

I think he mentioned in another add-on topic that all his app will be 4.7 compatible in next few days. I am awaiting for same as well. 😉 

7 hours ago, Randy Calvert said:

It's how long mySQL will hold a connection waiting for more requests from the same server.  (In your case, it's ALWAYS going to be the same since it's localhost.). Keeping the connection open longer means your server does not have to keep opening new connections to mySQL and instead can just use an existing one for longer.

Thanks. I still have question around it.. please help clarify and I appreciate the feedback. 

wait_timeout closed idle state non interactive connection. So basically let's say, connection is idle after 10 sec, it can be closed. 

My above error seems like active connection is 'forcefully' closed? How extended wait_timeout will help? Just asking because in past, I had it at 90 and I still had similar errors. 

Hi Ryan,

Following up - Is this resolved?

Thanks

20 hours ago, Randy Calvert said:

Try increasing the wait timeout to 60 seconds.  It's typically in /etc/my.cnf.  If the variable does not exist, add it.  

wait_timeout=60

Question is why to increase? To me no user is going to wait for 60 sec, staring at website... I would like to first understand, why query is taking 60 sec. 

I expect website to display in less then 10 sec at max and as fast as possible depending on latency, user location, etc. 

Slow query log is not helpful in finding slow queries cos it doesn't pin point issues for the logs which I shared. Anywhere else, I should look for?

2 hours ago, Miss_B said:

This seems to be server/host related. What is the wait_timeout value set to?

You can read more about this here:

https://dev.mysql.com/doc/refman/8.0/en/server-system-variables.html#sysvar_wait_timeout

 

That's set to 30 sec. 

1 hour ago, Randy Calvert said:

It's likely the mysql server on the host is too busy and the response timed out.  

I don't think so. 

-------- Performance Metrics -----------------------------------------------------------------------
[--] Up for: 69d 3h 48m 43s (347M q [58.114 qps], 8M conn, TX: 5865G, RX: 542G)
[--] Reads / Writes: 96% / 4%
[--] Binary logging is disabled
[--] Physical Memory     : 62.6G
[--] Max MySQL memory    : 25.0G
[--] Other process memory: 0B
[--] Total buffers: 14.6G global + 71.2M per thread (150 max threads)
[--] P_S Max memory usage: 72B
[--] Galera GCache Max memory usage: 0B
[OK] Maximum reached memory usage: 21.1G (33.71% of installed RAM)
[OK] Maximum possible memory usage: 25.0G (39.93% of installed RAM)
[OK] Overall possible memory usage with other process is compatible with memory available
[OK] Slow queries: 0% (365/347M)
[OK] Highest usage of available connections: 62% (94/150)
[OK] Aborted connections: 0.26%  (22371/8770303)
[!!] Query cache may be disabled by default due to mutex contention.
[!!] Query cache efficiency: 0.2% (653K cached / 299M selects)
[OK] Query cache prunes per day: 0
[OK] Sorts requiring temporary tables: 0% (175K temp sorts / 64M sorts)
[!!] Joins performed without indexes: 5410235
[!!] Temporary tables created on disk: 27% (6M on disk / 23M total)
[OK] Thread cache hit rate: 99% (94 created / 8M connections)
[OK] Table cache hit rate: 99% (837M hits / 839M requests)
[OK] table_definition_cache(4000) is upper than number of tables(1348)
[OK] Open file limit used: 5% (295/5K)
[OK] Table locks acquired immediately: 99% (17M immediate / 17M locks)

Spike is because I was running some longer queries. But overall not much longer running queries. 

Also no delays in acquiring locks:

Also plenty of CPU - so I don't see any resource constrain:

2 hours ago, Jim M said:

Sorry, we do not have a schema and relationship documentation.

It would be nice to have one. I hope in future, you consider it. Thanks.

On 7/9/2022 at 4:46 PM, sibomots said:

A little better query so the admin can focus on the `preference_key` to examine later:

select M.name, M.email, PP.notification_key, PP.preference
from core_members M, ibf_core_notification_preferences PP
where  PP.member_id = M.member_id
and exists (select P.notification_key
from core_notification_preferences P
where P.member_id = M.member_id and
 P.preference = 'email')

 

I was looking for something like this.. Thanks for sharing. 

On 7/9/2022 at 4:22 PM, Jim M said:

Thanks for posting!

Unfortunately, this issue is beyond the scope of our technical support. 👩‍💻

Our technical support is happy to help you with the Invision Community platform, but we're unable to help with things like server management, theme questions and modifications.

I've moved this to our Community Support area where other Invision Community owners will see it and help where they can.

Is their any IPS document  -  which shares DB mapping? 

I am seeing below consistent problems... trying to track down, which query is triggering.. any suggestions? I have a feeling, it's coming from download. 

2022-07-13T06:44:02.365652Z 8488613 [Note] Aborted connection 8488613 to db: 'xxxx' user: 'abc' host: 'localhost' (Got timeout reading communication packets)
2022-07-13T09:59:02.386169Z 8499671 [Note] Aborted connection 8499671 to db: 'xxxx' user: 'abc' host: 'localhost' (Got timeout reading communication packets)
2022-07-13T16:16:01.951564Z 8530536 [Note] Aborted connection 8530536 to db: 'xxxx' user: 'abc' host: 'localhost' (Got timeout reading communication packets)
2022-07-13T16:59:07.843974Z 8535076 [Note] Aborted connection 8535076 to db: 'xxxx' user: 'abc' host: 'localhost' (Got timeout reading communication packets)
2022-07-13T17:59:41.600576Z 8542128 [Note] Aborted connection 8542128 to db: 'xxxx' user: 'abc' host: 'localhost' (Got timeout reading communication packets)
2022-07-13T18:54:44.054694Z 8547447 [Note] Aborted connection 8547447 to db: 'xxxx' user: 'abc' host: 'localhost' (Got timeout reading communication packets)

Any help is appreciated.

AWS - through API is better option vs SMTP. With SMTP we noticed forum slowness, issues.

Hello,

Can you please add support for custom themes? I have one from @Ehren - Elegant and Deflection and both themes doesn't show correctly.

Hello,

Share some more details on Website user - 'active users in past xx minutes, Database, Server Specs, hosting type, etc. so others can help you. 

1 hour ago, brfcs said:

We no longer use the gallery, Blogs or Pages apps ... is it possible to remove them from from our recurring invoice?

I believe IPS team can - but if you remove, you end up loosing those addons all together even though you paid for it. 

I wish IPS staff can tell you otherwise. 

On 5/22/2022 at 8:45 PM, Michael.J said:

I will be looking into Commerce integration to include stripe support as well. But that's an extra $100 now and costs a lot more now for users who really just need basic Paypal support.

On 4/18/2022 at 4:20 PM, KsiążeHappy said:

Hi Michael, 

IPS Nexus - doesn't support donation tracking like you do. 

Donation doesn't support recurring donation, so I have to use IPS nexus. 

All we want is basic and simple:

1. Allow users to donate - Your app does.

2. Allow users to create subscription - App doesn't and hence forced to use IPS Nexus.

3. Allow users to donate through non PayPal options - Your app doesn't and hence IPS Nexus. 

So now, users are getting frustrated because their recurring donations don't show up in Donation Addon and their profile count. 

Also no another way to pay then PayPal. A simple integration with Stripe will open all the doors for you. 

I am not understanding, why it's so complicated to add Stripe Payment system and recurring donations in your addon. 

Nothing against you but I hope you understand my frustration which is the result of our users getting frustrated and it's management nightmare. 

I just hope after requesting for recurring donations for few years, you will add that in your add-on.