Tested this issue using reported version #59.0.3071.109 on Mac 10.12.5 and Linux Ubuntu 14.04 and was able to add the above attached extension and did not see any error message "This extension may have been corrupted."

@luokobe: Could you please upgrade your chrome to the latest version and check if you still face the issue?

Thanks!!

I am having the same problem on Windows 64 bit Chrome Version Version 59.0.3071.115 (Official Build) (64-bit).

I can't post the webstore page as it's only available for private testers. Sorry.

No errors when loading unpacked extension.

No errors on uploading to Web Store.

This has occurred on an update that does not materially change the extension very much.

When the testers try to download the extension, it appears as a download in the bottom left hand corner of Chrome, which it never used to, as a standard download. Then it is just marked as "This extension may have been corrupted."

#4 - is this happening on Windows as well?

Could it because the extension contains css\tab_content.css file with 0 size? Zero is divisible by 4096 and this is a known cause for the issue.

#5 Yes I'm only on Windows.

It also happens on Chrome OS. It seems that this happens when the extension gets an update.

Thank you for providing more feedback. Adding requester "sandeepkumars@chromium.org" to the cc list and removing "Needs-Feedback" label.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Uanble to reproduce the issue on Mac OS 10.12.5 and windows 7 using chrome M59 #59.0.3071.115 and canary M61 #61.0.3150.0.

Attached screencast for reference.

@Could someone from extensions team look into this.

Thanks!

Deleted: 736699.mp4 1.2 MB

	736699.mp4 1.2 MB View Download

For what it's worth, it seems to be Chrome 59 problem only.

The issue is not occurring for me with the same extension on Chrome Canary
Version 61.0.3150.0 (Official Build) canary (64-bit)

I can confirm the missbehavior under both Chrome OS Version 59.0.3071.113 and Version 60.0.3112.50 (Build officiel) beta (32 bits).

Steps to reproduce the problem under CrOS 59.0.3071.113:
1. Install the extension from web store
2. The extension will be disabled and says "This extension may have been corrupted."
« Repare » the extension makes it work for a moment, few minutes max, then back to 2. Same when « Remove » and « Install » again.

Steps to reproduce the problem under CrOS 60.0.3112.50:
I then shifted to ChromeOS beta while the extension was in the « may have been corrupted » state:
- Extension was still in that unusable state after rebooting;
- I « Repared » it but it went back to the « may have been corrupted » state the next minute (I couldn't « add » the bookmark I had started since I couldn't save it).
- Trashed the extension then re-installed it (Diigo Web Collector - Capture and Annotate 3.3.59) from the Chrome web store; reloaded the pages and could annotate for about 5 minutes. Then it went back to the « may have been corrupted » state.

I can reproduce this bug on chromebook Version 59.0.3071.113.

As soon as I select any text on the page, the extension disappeared and is shown as corrupted

Con confirm on ChromeOS Version 59.0.3071.91 (Official Build) (64-bit)

Repair works to re-institute a working version of the extension (3.3.60), but every time I have done so, it fails again later.

To test this, I repaired the Diigo extension on Friday -- as usual, it reappeared and worked as expected. I left it over the weekend, not using my computer at all, and came in on Monday to find that it had  been disabled, as usual, with the message 'This extension may have been corrupted.'

@hdodda,  I have reproduce this bug on Chromebook.  It can be successfully installed. However, after I logged in and try to highlight the page. The extension corrupted.

@hdodda, here is the video to reproduce this bug.

Deleted: bug2.mp4 3.6 MB

	bug2.mp4 3.6 MB View Download

We are still unable to reproduce the issue on Mac OS 10.12.5 and ubuntu 14.04 using latest stable M60 #60.0.3112.78 and canary M62 #62.0.3168.0 .

@luokobe-- Could you please confirm us if you can reproduce the issue in latest stable M60 #60.0.3112.78 .

Thanks!

I confirm this is still an issue in latest stable Windows

Version 60.0.3112.78 (Official Build) (64-bit)

@hdodda, here is a chrome book which can reproduce the issue.

Deleted: Desktop screenshot-chrome-browser.png 168 KB

	Desktop screenshot-chrome-browser.png 168 KB View Download

Deleted: Desktop screenshot-chromeversion2.png 290 KB

	Desktop screenshot-chromeversion2.png 290 KB View Download

Thank you for providing more feedback. Adding requester "hdodda@chromium.org" to the cc list and removing "Needs-Feedback" label.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

@hdodda, It seems I need to sign in the diigo account to reproduce the bug in Chrome book. Because the selection menu is only invoked after signing in.  

Hello,

I am experiencing the same issue with the updated Diigo extension. Everything has worked very well until, updating to the most recent version of Diigo: Version 3.3.66 (I am using a 2013 Chromebook Pixel running Version 59.0.3071.134 (Official Build) (64-bit). Once the extension is installed it works fine temporarily and then just disappears from the toolbar and needs to be re-enabled. This is extremely annoying since it is one of the extensions I am constantly using within the platform. Thanks.

@hdodda, 
1)We use the same package and test it in Chromebook through "Load unpacked extension". It works fine, no corruption issue.
  
2) If this package is uploaded to Chrome web store, then we install it through Chrome Webstorein Chromebook, corruption issue as in the video appear again.

There is a corruption issue for recent uploads to Chrome web store. We are working on a fix ASAP.

Awesome. Glad you found it. Will you post here when it's fixed? It already seems to be OK in Version 61.

The issue may not manifest on the dev channel, due to differences in how extensions are verified on the dev channel. :(

There's also a chance that there are multiple underlying bugs here - we're trying to sort out exactly is causing what.

I will post updates when available.

Thanks for looking at it. Appreciate it.

louk.../joel.../mail.../tuko...: Can you confirm that this is no longer occurring with the version of diigo uploaded on August 2nd? (The current version).

thom...: Can you confirm that your extension contains a file with a file size congruent to zero mod 4096? I.e. that the extension contains some file of size 0, or 4096 or 8192 or 12288 or so on...? If you post the extension ID I can use our malware detection tools to do this for you.

The server-side corruption issue that we discovered does not affect these extensions. Because the reports are that this is working in M61, it seems likely that these are duplicates of  crbug.com/720597  (the %4096 bug), which is fixed in M61.

Cheers my extension id on the Chrome Web Store is:

lplkcofhmlcehlkaponalfknpbnmnojm

I've looked through for files of 4096 but I must have missed one if you think that's what it is.

Thanks.

utils/storage.js has a size of 8192, so that is the problem. As a temporary workaround, if you can add a comment to the file to modify the size, that is the quickest path to your case being resolved.

I'm investigating the merge story for  crbug.com/720597 ; I think we're likely too close to the branch cut for a merge, though, given the existence of a workaround.

@Waff...  I still experienced the same crash issue with the August 2nd version https://chrome.google.com/webstore/detail/diigo-web-collector-captu/pnhplgjpclknigjpccbcnmicgcieojbh .  I am using an acer Chromebook as reported by Luo above. The current chrome version is 60.0.3112.80(64-bit).

Thanks joel, I am trying to repro.

Here is the video for the bug.

Deleted: AwesomeScreenshot-2017-08-04T18-02-49-125Z.mp4 2.1 MB

	AwesomeScreenshot-2017-08-04T18-02-49-125Z.mp4 2.1 MB View Download

@waff..  Let me know if you need any info.

Thanks for the video, that's very helpful! Confirmed the repro for Diigo Web Collector 3.3.69 on 60.0.3112.80 (ChromeOS). Digging into what is going wrong (and why)...

Great!  Appreciate it very much. 

Also repro'd on Linux.

[17290:17326:0804/115309.463792:VERBOSE1:content_verify_job.cc(230)] job failed for pnhplgjpclknigjpccbcnmicgcieojbh diigolet/images/search-csm.png reason:3
[17290:17290:0804/115309.463949:VERBOSE1:content_verifier.cc(143)] VerifyFailed pnhplgjpclknigjpccbcnmicgcieojbh reason:3

Reason 3 indicates a hash mismatch, but I think it's more likely that the issue is due to "search-csm.png" being requested but "Search-csm.png" being present in the archive:

renaming diigolet/images/Search-csm.png to diigolet/images/search-csm.png on my file system fixes the issue. I'm guessing this was not an issue on Windows because the filesystem is case-insensitive. Can the extension be updated to include search-csm.png rather than Search-csm.png?

Thanks very much! We will do it and report it back in a few hours and report it back.

BTW, if the package is loaded from "Load unpacked extension" in manage chrome extensions page of  Chrome browser (the 3.3.69 with the upper case Search-csm.png), it works. However, if it's installed from webstore, it triggered the issue. What's the mechanism behind this? 

The corruption verification is non-fatal if you load an unpacked extension - the idea was that it's too hard for developers to produce the verified_contents.json file that is used to check the file hashes (because (A) only Webstore has the key to do so, and (B) it requires you to know in advance the CRX ID, which you can't know from the unpacked form of the extension). So I think in that case the verification may be skipped altogether, or postponed until browser reboot, or something like that.

I was the 4096 bug, Thanks @waff for looking at this and sorting it out for me.

@Waff,  Thanks for the explanation.  We updated a new version with the workaround and it works on my Chromebook.  I have requested some other users to confirm the fix and  will let you know the results.

Thanks for your inspection! 

@Waff,  Now there are more confirmations and the issue was solved.

Thanks again!