sadams101

#1 above should say "topic", rather than post.

Thank you for the offer, but my solution outlined above did work. Here is a summary: 1) Add a new post to the forum; 2) Wait for 15 minutes or so (I think one of the cronjobs runs and updates the number of posts...not sure which one) I don't see a reason to try to find out the cause at this point.

PS - This doesn't seem to work all the time, but I noticed it did work in a few of my forums.

I found out some more info which I'd like to report. I've been watching the counts for each forum that I have, and have noticed that some have updated to the correct counts. This made me wonder why some updated, because the changes had nothing to do with me clearing the cache. I noticed that when a new topic was posted in a forum with an incorrect count, the total number of posts updated to the correct count. Why would this be? What caused the counts to be off in the first place? I suspect that when I ran the last update the counts were somehow changed to the incorrect low counts, and now when a new topic is posted this gets fixed. The reason that I believe this is because the date of when I ran that update seems to fit with some of my forums, which have been posted to since I ran that update.

That makes sense, but when I switch back to Redis will the cache be flushed, or should I do that manually from command line? What I don't want to happen is for the old values to start displaying again if they are still stored in the Redis cache...so would using this command after I switch caching methods ensure that this won't come back? redis-cli FLUSHALL

Thanks, I know how to disable the Redis cache, but is there a way to also flush the Redis cache, and is it safe to do this?

Thanks, none are running, but they did run and take several hours to complete. Unfortunately many of the counts are still way off for most the the various forums. Any idea what might cause this?

Although I've cleared the Admin's "Clear System Caches" multiple times after running the recount, this still seems like a cache issue. I just posted the prior screen shot a few minutes ago, and now I see that the "Celiac Disease Pre-Diagnosis, Testing & Symptoms forum update its posts--just a few minutes later--see below. Can you tell me how this info is cached. I am using Redis and perhaps it isn't clearing properly?

After the recount process ran a couple of the forums updated their counts, but most are still far off. Is there any command line script that can be run to force the recounts? Are archived posts included in these counts? LIVE SITE after recount: Strangely it looks like only a couple the different forums updated, but most did not.

After looking things over, it doesn't appear to be the case. The number of topics in each area closely matches between both sites. PS - and my moderators don't have any deletion powers.

I am running a recount now, thanks for letting me know where that setting went to...but both sites have the same percentage of archived posts (90%), yet the count is so far off between the two that it seems very strange to me. The only real difference in the two sites is the test site uses no caching, and it also does not use the built in cron job to update things. Here is the live site with count examples: Here is the TEST site with count examples: I will let you know how the recount process goes--it seems to take a while.

I am running 4.7.20 and my topic and post counts are inaccurate. When I compare my test site with my live site they are totally different. I no longer see the recount tool in the AdminCP, and I've tried reverting my skin to the default, but the inaccurate counts did not change.

I was investigating some strange log entries, and may have found a bug. Here is an example of the log entry with the domain in URL changed because I don't want bots hitting it: The log entry was triggered by a guest (Guests) Sunday at 06:03 AM The URL of page the error occurred on was https://www.myURL.com/pastleaders/?sortdirection=desc TypeError: Cannot assign string to property IPS\Dispatcher\_Front::$currentPage of type ?int (0) #0 /home/celiac/public_html/system/Dispatcher/Dispatcher.php(110): IPS\Dispatcher\_Front->init() #1 /home/celiac/public_html/index.php(13): IPS\_Dispatcher::i() #2 {main} BACKTRACE: #0 /home/myURL/public_html/init.php(1049): IPS\_Log::log('TypeError: Cann...', 'uncaught_except...') #1 [internal function]: IPS\IPS::exceptionHandler(Object(TypeError)) #2 {main} This led me to do some testing on this page, and I was unable to click on Page 2, or Page 3 and get the results to change, and I see that your site has the same issue (after going there just try clicking on the different pages of past leaders, and they results don't change): https://invisioncommunity.com/pastleaders/ Perhaps the "sortdirection" in the example link in the error is something that has been discontinued, but even the Jump To Page function does not work, and there is no way to display any older past leaders than those that appear on the initial page.

Since this is still an issue in the latest version, I am circling back to see if this is slated for a future update?

I am on the latest release 4.7.17, and only mentioned that upgrade as the point where I first noticed sitemap resource issues from my server. Please double check your sitemap code changes in that update. I suspect there is an issue.