Bug #6286
closedGroups pages not displaying at all!
Added by Marilyn Weber about 8 years ago. Updated about 7 years ago.
0%
Description
Katina Rogers reported:
"Groups pages are silently failing for me; the following pages will not load but display only an empty browser window, no error message:
https://commons.gc.cuny.edu/groups/digital-humanities-initiative/
http://commons.gc.cuny.edu/groups/
I'm using Chrome and the error occurred at 9:42am on Friday, 10/14. I don't need any follow-up or support, just reporting the error."
I'm also using Chrome and am seeing the same thing. Two screenshots attached.
Files
| DHIBlank.JPG (31 KB) DHIBlank.JPG | Marilyn Weber, 2016-10-14 10:13 AM | ||
| GroupsBlank.JPG (325 KB) GroupsBlank.JPG | Marilyn Weber, 2016-10-14 10:13 AM | ||
| Screen Shot 2016-10-14 at 11.18.46 AM.png (162 KB) Screen Shot 2016-10-14 at 11.18.46 AM.png | Matt Gold, 2016-10-14 11:19 AM |
Related issues
Updated by Matt Gold about 8 years ago
- Category name set to Groups (misc)
- Status changed from New to Assigned
- Assignee set to Boone Gorges
- Priority name changed from High to Immediate
Boone -- are you around to take a look at this?
Updated by Marilyn Weber about 8 years ago
Wait - they mostly work. Neither shows avatars.
Updated by Matt Gold about 8 years ago
I'm still getting failures on http://commons.gc.cuny.edu/groups/
Updated by Boone Gorges about 8 years ago
While I'm debugging this, if anyone is able to reproduce reliably, please do the following:
- Append ?bbgd=1 to your URL and reload the page. This will activate a temporary debug mode.
- Take a screenshot of error messages and upload it here
Updated by Matt Gold about 8 years ago
Hi Boone -- this is from the Groups page. Mac / Chrome
Updated by Boone Gorges about 8 years ago
This appears to be a bug in the latest version of Chrome. I'm continuing to investigate workarounds.
Updated by Boone Gorges about 8 years ago
I've got a workaround in place. Can I get confirmation that it's working? I'm still looking into root causes.
Updated by Matt Gold about 8 years ago
Yes, confirmed that this is working for me. thank you so much, Boone.
Updated by Boone Gorges about 8 years ago
- Target version set to Not tracked
The deepest cause I can be sure of is that Chrome is not receiving the end character of some chunk of the page, and is therefore refusing to render it. See http://stackoverflow.com/questions/29894154/chrome-neterr-incomplete-chunked-encoding-error/29990521 for one thread describing the issue.
This error can happen for many different reasons: a fatal error in an output buffers, Apache sending the wrong Content-Length header, improper gzip compression settings. The fact that the issue just started happening makes me think there may be a recent server change or a recent browser upgrade to blame, though it's not clear to me what it would be. I'll follow up with Lihua to rule out some possibilties.
I was able to find a workaround for the moment: disabling opcache. See http://stackoverflow.com/a/29970732. I have no idea why this fixes it. Ray, any thoughts?
Updated by Matt Gold about 8 years ago
- Target version deleted (
Not tracked)
Hi Boone -- two questions:
-- will this affect CBOX sites?
-- is this a BuddyPress-wide issue?
Updated by Matt Gold about 8 years ago
(I am seeing NYCDH groups with a problem)
Updated by Boone Gorges about 8 years ago
As described above, I'm afraid I don't have any idea about the root cause. I'll follow up if/when I have more info.
Updated by Raymond Hoh about 8 years ago
I'm trying to duplicate this on cdev, but I cannot. Tested on Chromium 56.0.2888.0 and Chrome 51.
Maybe this only occurs on production?
Updated by Raymond Hoh about 8 years ago
- Related to Bug #6310: Members of groups kicked out of the group including owners added
Updated by Raymond Hoh about 8 years ago
- Related to Bug #6313: group member no longer in group, can't rejoin. added
Updated by Raymond Hoh about 8 years ago
- Status changed from Assigned to In Progress
Boone did some investigating and found out that one of our memory caching instances was failing, which was probably the cause of all the other issues today (#6310, #6313).
To temporarily resolve this, we have disabled the instance causing the failure. We will relay this info to Lihua to see if he can run down the cause of the problem.
Updated by Boone Gorges about 7 years ago
- Status changed from In Progress to Resolved
Haven't seen this problem recur, so I'm closing.