[BUG] Instagram stories Graphql download issues - intermittent downloading

[donbeatz]

Describe the bug
When downloading instagram content. IF "Use graphql to download" is checked, both automation and manually triggered downloads works perfectly - except for users active stories which sometimes download and sometimes don't. (the 24 hour stories). Even when a user has a known story and a manual download is triggered it won't get that story.

IF "Use graphql to download" is unchecked, it will download those stories that it misses or can't download when using graphql, but after 2 - 3 profiles, instagram detects botting and you get hit with a "Instagram credentials have expired [400]: [Instagram] - targetusername [Stories]" indicating that it was the stories that tripped it.
Then it returns " [TIMELINE, REELS, TAGGED, STORIES, USERSTORIES, SAVEDPOSTS] DOWNLOADING IS DISABLED UNTIL YOU UPDATE YOUR CREDENTIALS"

If you update your credentials, you get another 2-3 profiles and then get the samething. So it seems graphql is needed, but it will randomly miss stories, both video stories and photo stories, seemingly at random, even with forced manual downloads. It just seems to have a real problem with some stories.

I have tried to set up a few different story formats on my throw away insta account to see if it was a resolution/tag/hyperlink/shared story issue and all of them downloaded using both download methods.

None of the timing or limits have been changed from default. Is anyone else experiencing stories only being skipped randomly?

Please complete the following information or replace the following text with data copied from SCrawler (click the top right info button in the main window, then the Environment button, then the Copy button, and paste the copied text here).
SCrawler v2025.6.12.0 (x64)
OS: Microsoft Windows 10 Pro (x64)
ffmpeg version: ffmpeg version 5.1.2-full_build-www.gyan.dev Copyright (c) 2000-2022 the FFmpeg developers
yt-dlp version: 2025.06.09
gallery-dl version: 1.29.7
cURL version: curl 7.86.0 (x86_64-w64-mingw32) libcurl/7.86.0 OpenSSL/3.0.7 (Schannel) zlib/1.2.13 brotli/1.0.9 zstd/1.5.2 WinIDN libssh2/1.10.0 nghttp2/1.50.0 ngtcp2/0.10.0 nghttp3/0.7.1 libgsasl/2.2.0

Oh and all of the cookies and tokens/ids are correct. Again everything works 100% of the time - apart from stories which work about 50% of the time.

Another clue I found:
When the stories fail to download any content, the Stories (user) folder in the targets profile on disk, its date changes to the current day of scanning. So if you know profile x has a story, you manually trigger a download. and it says its got everything, if you go to profile x's folder and look at the stories (user) folder where the content should be the content is missing, but the folder date has changed.
So it seems like it is trying to write data to that folder.

This seems to be happening across all profiles (about 20 im monitoring), and it seems that one specific days all profiles stories wont download, but other days they will (if they have stories).

Honestly have no idea what is going on and I can only code python so im out of my depth on looking at the source in git to trouble shoot.

[AAndyProgram]

Well done. Thank you for your issue.
So, step by step.....

GraphQL has been added for future purposes. It was completely debugged when it was added, but things may have changed over time. Currently I can't debug this algorithm just because Instagram has very sensitive ban algorithms. I can't risk my account. And registering fake accounts is very annoying because of their strong banning algorithms. I'm ready to continue this discussion on our Discord server.

Also make sure you read all the tips.

So please tell me is GraphQL more stable than the default?

Honestly have no idea what is going on and I can only code python so im out of my depth on looking at the source in git to trouble shoot.

These source code links may also be useful for troubleshooting:

• Default algorithm
• Launch function
• Main execution function
• Default universal parser
• Data extractor
• GraphQL functions

When the stories fail to download any content, the Stories (user) folder in the targets profile on disk, its date changes to the current day of scanning. So if you know profile x has a story, you manually trigger a download. and it says its got everything, if you go to profile x's folder and look at the stories (user) folder where the content should be the content is missing, but the folder date has changed.

Why are you manually checking this folder? Do you know about 'Feed'?

If you join our Discord server, please post a link to this issue in the help channel.

Join our Discord server: https://discord.gg/uFNUXvFFmg

[donbeatz]

Yes Graphql works best. It grabs every post/video on insta, it just sometimes misses stories. I've had graphql run for 3 months with 0 issues aside from the stories sometimes not getting downloaded, not even having to update credentials during that time. If its not enabled I can maybe get 3 profiles before my credentials expire. So its definitely good at what its doing, its just got this one weird bug.

Feed is good, but I was checking the folder to see if the were in there anyway. Also wanted to check the folders to make sure that maybe it wasn't a naming thing where it was using non English characters or too many characters. I know windows has a limit on file names but you can still create files the have longer names than it can handle and wanted to make sure this wasn't the issue. (It wasn't).

I was using instadownloader the other github project for insta scraping and it always gets stories, but it easily gets detected and my accounts have been nuked when using it - same sort of thing, it will get 3-4 profiles in and get detected, but when it does get detected it really gets nuked- like Instagram makes me verify my identity, change passwords, even send them a picture with a code they generate to verify im not a bot ect... So it seems like its not using graphql. If you read their issues on github, lots of users get the instagram 400 error, but it seems to be affecting people outside of America mostly, I suspect Instagram has slightly different verification/checks methods by region. If I don't use graphql on SCrawler I need to refresh credentials, so something inside graphql is getting around that check which is good, but yeah not sure why it only sometimes grab stories.

I will do some more sleuthing later in the week, I will check if its a region thing, by trying to grab stories from accounts in other regions. (ie korea, europe, south america, australia, merica,). It might be a date/time related bug where if you are in Australia and its 10 o clock and you are scraping a US account which is almost a day behind the instagram api would think its more that 24 hours old (ie if the story has been published for 10 hours when you go to scrape it and there is a 15 hour time difference it would be more than 24 hours old. Is the way the instagram api is being accessed sending UTC or is it sending local time of the machine scaping and the instagram api thinks that its trying to view an expired story?).

Ill give a go reading the VB source code but its way out of my wheelhouse hahaha.

[donbeatz]

So quick update.
Its not a timezone issue, as best as I can tell its a region issue and how the instagram api works in different regions. So each time I put in fresh credentials, I can get 3 or so story scrapes consistently, but then it becomes spotty. I tried vpns but had limited success.

So the log puts out this for every instagram account im scraping. It seems that as soon as whatever triggers it, all stories stop scanning and return (7).

2025-07-10 07:39:33: [Instagram] - andrea_leo_couture: (7) Download skipped (Stories)
2025-07-10 07:41:34: [Instagram] - hannahkenerly: (7) Download skipped (Stories)
2025-07-10 07:43:28: [Instagram] - louise_thielsen: (7) Download skipped (Stories)
2025-07-10 07:45:28: [Instagram] - tpkjoler: (7) Download skipped (Stories)

It grabs everything else on the profile but not the stories (and I did check that there were stories published when this scan was run).

I scanned the code in the links you sent, but couldn't see where it was generating the "(7) download skipped stories" error code from.

Could you update the story grabber for instagram to spit out more data to the log. Maybe have:

• Stories not enable on this profile (user profile) skipping (8)
• No stories for (user profile) (9)
• API error for (user profile) stories (10)
• Token issue on (user profile) stories (11)
• Rate limit hit on (user profile) stories (12)

That way it might be easier to diagnose the issue. I suspect its an API error or something.

Is there a way to see a more detailed log, specifically looking at what Instagram's API is saying back to scrawler on each request?