Service Fabric Explorer just loads blank white page and requires refresh to load

[MikeChristensen]

The first time I load SFX, I just get a blank white page. I have to hit CTRL+F5 one or two times to get it to load. When it doesn't load, I see this in my console:

Browser info:

Google Chrome	74.0.3729.169 (Official Build) (64-bit) (cohort: Stable)
Revision	78e4f8db3ce38f6c26cf56eed7ae9b331fc67ada-refs/branch-heads/3729@{#1013}
OS	Windows 10 OS Build 17763.503
JavaScript	V8 7.4.288.28

[jeffj6123]

Hello,

What version of service fabric are you running?

[MikeChristensen]

Current fabric version 6.4.658.9590

[jeffj6123]

Are there specific pages of the site where this happen or does it happen for all pages? Also how often does this happen?

[MikeChristensen]

I've only tried this for /Explorer/index.html#/ since that's what I have saved in my bookmarks.. It happens 100% of the time. I go to that location, I see a blank white page, I hit CTRL+F5 to refresh, and the site continues to work from that point on. Next time I go to the site in a new browser session, same thing.

[jeffj6123]

Do you have the same thing happen if you visit just :19080?

[MikeChristensen]

Yes. Same thing.

[jeffj6123]

Do you see this with any other clusters you view or just this one?

[MikeChristensen]

Every single one.. We have a dev and staging environment setup, and I've setup several of my own test clusters. The issue happens on all.

[tudorsibiu90]

I have also experienced this for months. Shame on me for not reporting it, but it is so annoying...

[jeffj6123]

@MikeChristensen , @tudorsibiu90 do you know which versions you started seeing this happening in? also @tudorsibiu90 are you getting the same error in your developer console?

[MikeChristensen]

Not sure, I've been using SF for probably a year or so now, and it's always happened..

[jeffj6123]

Hey, So Service fabric 6.5 is going to be rolled out shortly and could potentially fix your issue. So I would be curious to see if you continue to have this issue once you are on 6.5 and if so we can follow up in official channels to set up a way to further diagnose this.

[MikeChristensen]

Sounds good!! I'll try again once 6.5 is released..

[olivergrimes]

I'm also having this issue in Chrome, multiple times a day on multiple clusters.

The workaround for me seems to be:

F12/Dev Tools > Application > Clear Storage > Clear Site Data

That does the trick every time. Hopefully it's already fixed in the update mentioned.

[MikeChristensen]

@olivergrimes CTRL+F5 should work as well..

[olivergrimes]

@MikeChristensen That has never worked for me - still doesn't!

[MikeChristensen]

6.5 made things WAY worse! Now this happens no matter what, and a CTRL+F5 won't fix it. SFX is completely broken now. Why was this update ever released?

[jeffj6123]

@MikeChristensen I am sorry for the issues you are hitting and are you still getting the same error or a different error? Also you can try clearing the site data because emptying the cache and hard reload does not clear where the data is stored related to aad, at least in chrome. This is why the trick oliver posted above works because that will clear the localstorage.

[MikeChristensen]

Yup, same error as the stack trace above (cannot read selectTreeNode of undefined). Confirmed clearing local storage works.

[tudorsibiu90]

this is only happening in Chrome as far as I can see and it is so frustrating I cannot even begin to describe it. This has been an issue for a year now as I have also reported it multiple times. I cannot believe it's still not fixed. for me, there is a last "authorize" call going on which gets automatically canceled for some reason at the very end of accessing the explorer page.

[MikeChristensen]

I've been using the new Edge (Which is also based on Chromium engine) and it repros there too.

[mhofkens]

same here... this has been an issue since the beginning of service fabric. Service Fabric version 6.5.641.9590

[jeffj6123]

Hey, As an update on this issue the fix that should hopefully remove these issues was not in 6.5 CU1 but is in for 6.5 CU2.

[MikeChristensen]

That's good to know! Is there a release date planned? This bug is driving me nuts!

[dannyfhalpotia]

Any update on expected release date?

[jeffj6123]

Sorry for a very slow update. But 6.5 CU2 has been getting rolled out currently which should hopefully resolve these issues.

[MikeChristensen]

@jeffj6123 Can you tell us the build number for this version? (6.5.???.????)

[jeffj6123]

The version would be 6.5.658.9590

[cbogles]

Hi, it seems the latest release solved the issue for us. Now a simple refresh is enough, thanks!!

[jeffj6123]

I am going to keep this issue open a little longer to see if anyone else continues to have issues @MikeChristensen @olivergrimes @tudorsibiu90 If you guys can report back when you have upgraded aswell.

[mhofkens]

Our dev cluster is running 6.5.658.9590 and works great. Thanks for fixing this.

[olivergrimes]

@jeffj6123 - Our clusters are now on 6.5.658.9590. Looking good so far! Thanks for the fix 👍

[tudorsibiu90]

@jeffj6123 Same here. It works now. Thank you!

[MikeChristensen]

@jeffj6123 Also confirmed this works. You probably have to clear your browser cache after the upgrade thought.

[jeffj6123]

I will be closing the issue now. @MikeChristensen that could be true, it shouldnt need to be cleared but that extra precaution could help.