aComAdi Posted October 17, 2020 Share Posted October 17, 2020 Hi, I am aware that this is a very broad description of a problem - but that's because I could not narrow it down so far. I just wanted to see if others have made that exprience or if this is even a known issue. On a regular basis, when logged into Processwire, the CMS and the front-end becomes nonresponsive. What happens is usually: User works for a while logged into the CMS and has another tab open, where she previews changes At some point, during a refresh/load of a new page IN THE CMS, the site is in an endles loading loop. When this happens, reloading the page, klicks on menu items, nothing responds. Closing the tab and and reopening in the same browser again results in endless loading loop (and eventual time out). When this happens, the front-end also cannot be loaded anymore in that browser. It's noteworthy that: Opening the site in another browser works Logging into Processwire in another browser works This happens regularly across various setups: Different browsers (Chrome, Firefox - haven't seen on Safari yet, because that's usually the last one I use to keep edition after Chrome and Firefox are 'locked') Different servers Different PHP versions (most installs are on 7.2.) Different versions of PW. I had it on 3.0.128 up to 3.0.165 now. Having site in debug mode or not Having cache activate or not Common to the installations is usually: Tracey Debugger RepeaterMatrix Rockmigrations Has anybody experience this? I haven't had any complaints from the clients yet - but that doesn't mean it does not happen to them. Internally this does happen regularly. Almost during every longer work session. Thanks for any insight. Adrian 1 Link to comment Share on other sites More sharing options...
dragan Posted October 17, 2020 Share Posted October 17, 2020 First things I'd do: check PW logs check server error logs (Apache / PHP) inspect in the browser (JS errors) If you don't find any hints that way, you could see if it's a CSRF problem, or something to do with sessions. Switching to SessionHandlerDB can help (a core module). In one of my setups I had to put these two config values to false, when I encountered similar problems: $config->protectCSRF = false; $config->sessionFingerprint = false; 1 Link to comment Share on other sites More sharing options...
teppo Posted October 17, 2020 Share Posted October 17, 2020 I've had similar issues with Chrome (probably because that's the only browser I really use) many times over, though not sure if it's ever occurred with a ProcessWire site. From your description it seems quite likely that "something" in the browser itself gets stuck — I'm really not an expert here, but again it's happened a number of times for me, and only things that seem to help are a) waiting until it goes away (could be a long wait), or b) rebooting the browser. Pretty much just guessing here, but I'd suspect either some sort of JavaScript issue, a problem with a browser extension, some sort of network issue, or perhaps some piece of software installed locally. Ad blockers, for an example, are known to cause browsers to freeze, but typically only under some very specific conditions. (I've also helped clients debug some pretty weird issues cause by firewalls or virus scanners.) Probably not very helpful, sorry; if you can see any console warnings or such please let us know, that'd make debugging much easier ? Link to comment Share on other sites More sharing options...
flydev Posted October 19, 2020 Share Posted October 19, 2020 Does the server(s) setup include NGINX ? A server/php setting that could cause the browser hanging on a script instead of terminating it ? Hard to tell without logs. Link to comment Share on other sites More sharing options...
aComAdi Posted October 20, 2020 Author Share Posted October 20, 2020 Hi, Sorry for the late reply. First case of Corona in the company. All sorts of emergency actions necessary. I will check @dragan's suggestions. @flydev ??Relatively sure no NGINX involved in any of the incidences this occured. Will report back if I find anything. Link to comment Share on other sites More sharing options...
aComAdi Posted February 26, 2021 Author Share Posted February 26, 2021 I figured, I will post the update: Turns out the culprit is Tracey Debugger. Once turned off, this phenomenon does not happen anymore. Luckily we only need it during installation and development of new modules really. All editors are now working with Tracey off and the problem has disappeared. 1 Link to comment Share on other sites More sharing options...
dotnetic Posted March 1, 2021 Share Posted March 1, 2021 On 2/26/2021 at 6:48 AM, aComAdi said: I figured, I will post the update: Turns out the culprit is Tracey Debugger. Once turned off, this phenomenon does not happen anymore. Luckily we only need it during installation and development of new modules really. All editors are now working with Tracey off and the problem has disappeared. Expand This could be because TracyDebugger loads Logfiles (by default?), namely Tracy Logs and ProcessWire Logs. If these files are large, it could take a while and make the site unresponsive. Try to disable them in Tracy's panel selector and see if that helps. 1 Link to comment Share on other sites More sharing options...
adrian Posted March 2, 2021 Share Posted March 2, 2021 Hi @aComAdi - sorry about the problem with Tracy. You know, this is one of those strange things I have seen with Tracey on and off over the years. I actually just came across it again today, for the first time in a long time and @Robin S just PM'd me about it as well. In his case, it was the Field List & Values section in the RequestInfo panel, which once disabled fixed it for him. For me it's when an image called on the frontend is not available. The weird thing I've noticed in the past is if I load all Tracy panels, then the problem goes away. I'd like to finally get this sorted out though. I thought I was the only one ever seeing it ? I'll try to take another look soon, but I'd like to know if disabling the Field List & Values section fixes it for you, or if it is related to a missing image. Thanks for your help in narrowing this down. 1 Link to comment Share on other sites More sharing options...
aComAdi Posted March 5, 2021 Author Share Posted March 5, 2021 No worries. We greatly appreciate Tracy. I will work on a client site over the weekend and test these variations. Link to comment Share on other sites More sharing options...
aComAdi Posted March 7, 2021 Author Share Posted March 7, 2021 First report: Been working for an hour with Tracy on (iMac, Chrome Version 88.0.4324.192, PW 3.0.165, Tracy ). In my case it's not connected to images. At the time that the 'lockup' happened, I had both the backend and frontend open in the same browser. Reloaded a page (with almost no content whatsoever) to test something. Then the endless load started. No missing images on this page. Note, that I currently have three bd-dumps on this page: 1 array, two booleans. Link to comment Share on other sites More sharing options...
aComAdi Posted March 7, 2021 Author Share Posted March 7, 2021 Next lockup with log panels disabled. Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now