PJStalls

Hi, As a part of a daily routine we monitor the https://invisioncommunity.com/release-notes/ page to see if there has been any updates to the forums. When checking the page it would seem like the newest release was 4.5.4.2 dated 04/22/2021 with a warning triangle indicating this is a security patch and everyone is advised to upgrade as soon as possible. The thing is that we have already updated to 4.5.4.2 quite a while ago. IIRC 4.5.4.2 (or was it 4.5.4.1?) was fixed by going to "Support" -> "Something isn't working correctly" to sort it. Has this just been merged into the full download in the client area? Is this supposed to be 4.5.4.3, only mislabeled? Have noticed a couple of times there's a lag between the release notes page and the "System" -> "Applications" -> "Check for Updates", and also the release notes page has had its newest entry overwritten in stead of becoming a new entry in the list, so it can be difficult to really get the full picture of what's new. As far as I can tell only the date has been changed on that page. It might seem a bit hastily, but it would be great to get an update if there was a patch or not. Will start keeping the full download zips to have the possibility of doing a full diff if the situation should arise in the future, but as they say, hindsight is always 20/20. In advance, thanks! 🙂

Hi, Tried to download todays patch and run through the normal updating procedures (using manual mode by downloading the zip file), but after uploading and clicking continue I'm met with the following message: Typically the updating process is supposed here to go to a final step with a progress bar where it's rebuilding language files, themes, etc which has not happened. Clicking continue again from this menu changes nothing. If I open another tab and go to Support -> Something isn't working correctly, it does not warn about the patch anymore and says "Running latest version". From the error message it would sound like I'm trying to let it upload the files via ftp/sftp, but I'm clicking the manual button going via /?app=core&module=system&controller=upgrade&manual=1 Thanks.

Holy ... Not sure how I managed to miss that one. Had gone back and forward reading through stuff and every link kept dangling the 10 000 limit. Oh well. Would it be possible to get i.e Amazon SES and only use it as a SMTP relay and keep MX at google? And then add Amazon SES IP to SPF (not sure about DKIM)? Anyone knows if it's reliable or is it troublesome IP-reputation wise using SES? I tried to set up a cheap server with WHM/cPanel on digitalocean just to try to send some of the forum emails via self hosted, but all the larger email providers sent back a big "nope" saying ASN was blocked and whatnot. If anyone has any pointers or tips for services how to get this sorted, it'll be greatly appreciated 😃

Hi Nathan! We have three licenses, where the third one is only used by the forums.

Yeah, that's also the numbers I've been seeing, but we are incapable of sending more than literally a couple of hundred emails before getting blocked for a day.

Hi, We moved to Google Workspace a couple of days ago after having self hosted emails on a previous setup via WHM/cPanel, but it's been a right nightmare since. Have - Set up SMTP relay settings and followed Googles documentation (sending emails are not a problem via IPB, it's just a quota issue) - Set up SFP and DKIM in case that would help - Cancelled the trial period as it was suggested as an option to increase limits - Pre-payed for a month (3 licenses) as that also was suggested as an option to increase trial limits We're not sending out any type of bulk mail. The only mails being sent out are user registration, password reset, digest, etc. When looking through different help documents online, I keep seeing 10 000 being mentioned as limits, but we're getting error messages (see below) after sending somewhere among the lines of 150-200 emails (according to the email statistics page, we need to send around 250-300 a day unless it's possible to cut down system emails like "Your registration is complete!" and otherwise, but it would seem like a drastic move) Does anyone have any tips (or know what the quotas are) on how to get this resolved? > DATA 550-5.7.1 Daily SMTP relay limit exceeded for customer. For more information on 550-5.7.1 SMTP relay sending limits please contact your administrator or visit 550 5.7.1 https://support.google.com/a/answer/6140680#dailylimit message-id-here- gsmtp In advance, thanks

1) VB was installed at example.com/forum/, singular form 2) IPS is being installed at example.com/forum/, (same location) 3) No changes has been made in furl.json I did check out that guide (thanks for linking it!), but since the site has a custom CMS in the root I think the problem might lie within the ".php" location block. Don't really want to send "every file not found with the php extension" to the IPS index.php since the CMS is a huge part of the site and has tons of scripts in it. Going to showforum.php does give the server 404 errordoc and not the IPB 404, which is why I was leaning towards this being some issues with the php location block. Perhaps I need to test more and see if it's possible to have two location blocks for php files and just set up fcgi twice. I.e something like location ~ [^/]forum/\.php(/|$) { } location ~ [^/]\.php(/|$) { }

Thank you for your input Paul. After numerous of hours of testing I realised I'm probably asking the wrong questions. Making a copy of the default forum index.php and renaming it to i.e showforum.php actually made the redirect work, so I guess it's an issue of me misunderstanding what takes priority in the nginx config in the forum location vs the general .php location where fastcgi is set up. So the question is rather: does anyone have an nginx config with php-fpm they are using , or if anyone can see if there's anything that can be done with the current config to fix this? Worst case scenario I can just create copies of index.php to all the different names i need VB redirects from, but would prefer to have it the proper way. server { listen 443 ssl; server_name sitename; ssl_certificate /etc/nginx/ssl/sitename.pem; ssl_certificate_key /etc/nginx/ssl/sitename.key; root /home/username/chroot/public_html; index index.php; auth_basic "Test"; auth_basic_user_file /home/username/.htpasswd; location / { try_files $uri $uri/ =404; } location ~ [^/]\.php(/|$) { #if (!-f $document_root$fastcgi_script_name) { #return 404; #} fastcgi_split_path_info ^(.+?\.php)(/.*)$; fastcgi_pass unix:/var/run/php7-fpm-username.sock; fastcgi_index index.php; fastcgi_buffer_size 128k; fastcgi_buffers 256 16k; fastcgi_busy_buffers_size 256k; fastcgi_temp_file_write_size 256k; fastcgi_send_timeout 300; #abtest for 502 fastcgi_read_timeout 300; #abtest for 502 fastcgi_param HTTP_PROXY ""; # Mitigate https://httpoxy.org/ vulnerabilities fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name; include fastcgi_params; } location /forum { #rewrite ^/forum/forumdisplay.php(.*)$ /forum/index.php last; # did not work . threw 404 try_files $uri $uri/ /forum/index.php; } error_page 500 502 503 504 /50x.html; location = /50x.html { root /usr/share/nginx/html; } location ~* \.(jpg|jpeg|gif|png|css|js|ico|svg|webp)$ { access_log off; log_not_found off; expires max; } # ... loads of custom CMS locations after this }

Hi, In the process of converting a couple of vBulletin 3.8 forums and have some troubles getting old URLs to redirect (i.e forumdisplay.php, showthread.php, etc). The conversion itself is successful and all the data is there and everything else is working as intended. Some potentially relevant notes: * The converter has both been ran on a fresh 4.4.10 and 4.5.2 installation with the same result * Friendly URL tab in "System->Settings->Advanced Configuration" in ACP does not list anything related to old VB URLs. Is there any place I can check to see if the redirects are in play, or is that done behind the scenes by the converter? * This is on a new server and I just moved over the VB database, not the files. Are any of the VB 3.8 files supposed to be in the invision folder to make this work? * I can see in the database that the converter has inserted the rows that has the redirect info, so it seems to know whats what * nginx config has the following for the forum section: location /forum { try_files $uri $uri/ /forum/index.php; } If anyone has some tips or could see what could be wrong here, any help is appreciated! Thanks!

Edit: After some more testing working my way back version for version, it would seem like 4.5 supports up to elasticsearch 7.6.2. Every release above throws the errors in the original post.

Hi, Is anyone running elasticsearch 7.9 with 4.5.1? Could have sworn I read somewhere that it was supported and the files claims MINIMUM_VERSION = '6.0.0', UNSUPPORTED_VERSION = '8.0.0', but on a test install I'm getting the following errors whenever it's being used. This is when looking at all activity (/forum/discover/) Array ( [root_cause] => Array ( [0] => Array ( [type] => x_content_parse_exception [reason] => [1:72] [bool] failed to parse field [should] ) ) [type] => x_content_parse_exception [reason] => [1:72] [bool] failed to parse field [filter] [caused_by] => Array ( [type] => x_content_parse_exception [reason] => [1:72] [bool] failed to parse field [should] [caused_by] => Array ( [type] => illegal_state_exception [reason] => expected value but got [START_ARRAY] ) ) ) Backtrace; #0 /home/site/chro/public_html/forum/system/Content/Search/Elastic/Query.php(1123): IPS\_Log::log() #1 /home/site/chro/public_html/forum/applications/core/modules/front/discover/streams.php(360): IPS\Content\Search\Elastic\_Query->search() #2 /home/site/chro/public_html/forum/system/Dispatcher/Controller.php(101): IPS\core\modules\front\discover\_streams->manage() #3 /home/site/chro/public_html/forum/applications/core/modules/front/discover/streams.php(57): IPS\Dispatcher\_Controller->execute() #4 /home/site/chro/public_html/forum/system/Dispatcher/Dispatcher.php(152): IPS\core\modules\front\discover\_streams->execute() #5 /home/site/chro/public_html/forum/index.php(13): IPS\_Dispatcher->run() #6 {main} Searching gives same error. Backtrace: #0 /home/site/chro/public_html/forum/system/Content/Search/Elastic/Query.php(1123): IPS\_Log::log() #1 /home/site/chro/public_html/forum/applications/core/modules/front/search/search.php(700): IPS\Content\Search\Elastic\_Query->search() #2 /home/site/chro/public_html/forum/applications/core/modules/front/search/search.php(86): IPS\core\modules\front\search\_search->_results() #3 /home/site/chro/public_html/forum/system/Dispatcher/Controller.php(101): IPS\core\modules\front\search\_search->manage() #4 /home/site/chro/public_html/forum/system/Dispatcher/Dispatcher.php(152): IPS\Dispatcher\_Controller->execute() #5 /home/site/chro/public_html/forum/index.php(13): IPS\_Dispatcher->run() #6 {main} Any tips would be appreciated 🙂

Hi all, Working on a vBulletin conversion and was wondering if anyone had any experience with some potential file permission issues. Essentially, the forum is able to write files and folders as intended (everything from when i upload a new avatar/cover, writing templates to filesystem, a new skin I'm working on is able to create the folders and files for css_built_X, etc etc), but some functions typically related to read access aren't able to run. For instance: * Since this is on a test server I've disabled email functionality via SELinux to prevent accidentally sending out mails. Have done some actions (i.e reporting a post) and now the mail error log reports multiple lines with i.e filesize(): stat failed for /home/sitename/jail/public_html/forum/datastore/acpNotifications.xxxxxxxxx.php. Obviously I did expect errors since emails are disabled, but this indicates the forum doesn't have read access for the file in question (actually there are multiple from the same folder. failedMailCount.php, email_plaintext, etc). Uploading a file with the code below in the forum folder (having the same group/owner/permissions as the rest of the forum files and also double checked it's the same path as in the error) and opening it in the browser reports the correct file size without any issues. <?php echo filesize('/home/sitename/jail/public_html/forum/datastore/acpNotifications.xxxxxxxxx.php'); ?> * Trying to enable cronjobs gives the following error: In order to enable this setting you must make the /home/sitename/jail/public_html/forum/applications/core/interface/task/task.php file executable (i.e. chmod 0777). Looking in the code it seems to fail returning the status of is_executable() on the file. Once again uploading a file to the forum trying the same function on the same file, everything works as expected. Other notes: * nginx 1.17.10, php 7.4.6 (php-fpm) * Have double checked all files and folder has the correct owner and group * Everything in Support -> System Check reports as working * Misc log files on server does not report anything regarding the errors (audit.log, messages, nginx, etc) Leaving some notes below regarding SELinux and php-fpm. If anyone has any ideas on what to try, feel free to let me know! In advance, thanks ___ SELinux was set up as follows php-fpm has