Sporadic failure - "Error: WDE0053: Internal wixData error: WDE0055"

Hi All.

I observe sporadic failures while reading from collections.
Usually I get it like “Error: WDE0053: Internal wixData error: WDE0055: Failed to parse server response.”.
I’ve been in contact with WIX support for 3 weeks, but they basically say now that till I’m the only user with this sporadic issue, it is not a case and they won’t forward it to devs.

Does anyone else experience a similar issue?

Thanks

1 Like

Hi,

I have a scheduled task that runs through a collection every night updating statistics. Nearly every night I log one or two WDE0053/WDE0055 errors, always on an update (await wixData.update(…) ) within an async function. The failing item will work perfectly the next night, but then fail on another random item.

It’s almost as if there is a timing issue between the JS promises and the server replies when doing a series of updates in a loop. Definitely something for the developers to look at.

Cheers, Kevzone

In many cases, this is an infrastructure issue (request could not establish connection to server, it terminated unexpectedly, etc.). This is also the error you would get if some of Wix infrastructure is down.

It seems these errors are coming up more frequently for users, so we’re looking into it. Thanks for asking.

I am also experiencing this sporadic failure. Please let us know how we can get stability when saving to database collection.

Just throwing it out there that I’m also seeing this issue, also seemingly from updating a collection inside an async function. Looking forward to an update!

If your item amount is over 200000, you will get this easily

The WDE0053 has little to do with “over 200000”. That’s another issue (frontend timeout or db-token timeout). WDE0053 can, as Marlowe stated, pop up (even during a single .get) occur due to infrastructural problems. Good news is the WDE0053 usually doesn’t take long to go away. So when you do a .get or a .query or whatever, it is good practice to implement a retry when a WDE0053 is received : a loop with a max of 5 retries and 150 to 200 ms in between every retry. It usually works after 1 or 2 retries.
Hope this helps.

I agree that apparently it is not related to amount of data in the collection. I hit this error while quering collections with 10 items.

I think it is not related to front end. First because we are able to log the error at backend. Second because people report (above in the post) that same error happens in scheduled tasks which have nothing with front end in my opinion.

Regarding adding retries, I think it is not reasonable to add retries to 200 different queries I have on the backend for different stuff. Also add delays at WIX backend? Not sure it is good idea. If it is known issue, why not adding the retry with proper delay handling at infrastructure level?

However I think the thread regarding this issue will end up with retries suggestions (it is not first forum reply suggesting to handle the failure this way) and will not get the needed attention, as it happens with other similar hard to catch sporadic issues on the platform…

" If it is known issue, why not adding the retry with proper delay handling at infrastructure level? "
Yes, agree, it doesn’t belong at app-level. But it’s the next best thing we’ve got so far. Some of us, in an effort to make something more robust, are shifting away from written out queries per action and have started to develop a more generic framework , with only a query definition calling a generic, solid query handler with all that stuff built in. But retro fitting that could be costly, I know.

Giri, so eventually this is workaround and not a solution. Each developer will have to implement his own “robust framework” to avoid the WIX bug. Do I get it correctly?

To solve your issue on your own, you may need to do some comparison.

By reducing the amount of data called, check permission, filter properties(remove one and try), although it seems irrelevant, this is the only way you can try figure out yourself.

Sometimes wix just return “internal bug”, no other hints, so it really make user confuse.

OK, we did some talking. Read this : https://www.wix.com/velo/forum/community-discussion/vote-a-robust-data-handling-framework

WDE0053: Internal wixData error: WDE0055: Failed to parse server response.

I started getting this today. Initially I thought it is some server downtime somewhere. But it has happened for almost a whole day today. I know that I haven’t changed anything in my code to make this happen.

Me too. It started last night for me. I contacted Wix support twice, but until now no updates. One of the agents said it must be my code, but even when I created a new blank site with a fresh new account, I still received the same error. So it can’t be my code. It’s not only happening on one website, I receive the exact same error for all of my websites.

Check this out:
https://support.wix.com/en/article/5-february-2022-503-error-on-live-sites-in-taiwan-and-hong-kong

It says, it has been solved. When really it was not

and I want to add something, when i used vpn it works fine

Hi @haydibaswedan and @siliconvalleywoodtur This issue should now be resolved. Can you please confirm?

Getting this error as well. Has anyone figured out a solution for this? Been speaking with support for days with no resolution

Hi Joy,
Did this get resolved for you? If not, please provide the support ticket # here so I can pass it to our Data team to investigate. Thanks!

Velo team is still working on it. Got a response from them this morning but still waiting on a resolution. Request 311284342

Got it. Thanks!

An FYI (FWIW) to anyone seeing this error–

I just began to see this error when updating Wix collections from HTTP Fetch data returned from Google spreadsheets. I found that the error arose performing WixData.bulkSave when the content array for the bulkSave was empty. In the past, I believe it did not error in such a case; rather it returned empty results.

So, I have added a test to ensure the content array submitted for bulkSave is not empty. This has solved the error problem in my case.