# Seeing kopia connect crash having large number of epochs **URL:** \ **Category:** General Topics\ **Created:** [August 1, 2023, 4:14pm UTC](https://kopia.discourse.group/t/seeing-kopia-connect-crash-having-large-number-of-epochs/2189 "2023-08-01T16:14:13Z")\ **Posts on this page:** 1\ **Showing post:** 11
**Author:** ![jit010101](https://yyz2.discourse-cdn.com/free1/user_avatar/kopia.discourse.group/jit010101/32/785_2.png) [@jit010101](https://kopia.discourse.group/u/jit010101)\ **Post date:** [September 21, 2023, 8:11am UTC](https://kopia.discourse.group/t/seeing-kopia-connect-crash-having-large-number-of-epochs/2189/11 "2023-09-21T08:11:29Z")
> [@sgw](#): > > Tried latest development build from github. fails. Well it worked on the initial setup for me with the latest build - then it stopped connecting after rebooting the repository server once. And here you are having these issues on multiple clients … not sure if I’d have the patience to stick trying … that again. > And there are way bigger repos than mine, so I can’t imagine that the number of snapshots could break things here. This puzzles me too … but you can easily throw any amount of movie collections towards something like Kopia - so that itself is no real flag as how reliable it is in reality. For example my use case does introduce a Server solution like the Repository Server - they could all be using Rclone and S3 → that would offload a big chunk of stability on the Server side towards those solutions (until you have to restore e.g.). By technical theory it all sounds like Kopia could be something great - the code and tests didn’t look to bad too … and likely early experience wasn’t to bad too … maybe there’s a lack of attention or others just don’t bother? When I moved to test it (kopia 0.14.1) I immediately uncovered bad (=horrible) documentation, even on command-line, broken connect, non-streamlined commandline-arguments some weird design decisions and maybe an overly complex system that just isn’t stable enought - speaking from experience that implies a lack of real world testing and practical sanity in summary - before versions are released. (e.g. removing http support - given that every single reverse proxy does usually build on it by layering - one very simple example that works great in practice is for example restics rest-server) That is ALL written without doing even a SINGLE snapshot - having no data at a fresh repository. All in all if feels unpractical - and given I read about corruption and suffered the same connect errors without doing any snapshot / backup at all - that’s something really really off-turning (and I’m usually really subborn in trying to get shit running). What I’m speaking about are these issues - all mentioning that they **can’t even connect until they fix it on the file-system layer** : > [@Kopia repository 'corrupted itself' (unable to add to XXX index blob cache: context canceled)](https://kopia.discourse.group/t/kopia-repository-corrupted-itself-unable-to-add-to-xxx-index-blob-cache-context-canceled/1766): > > I had a working Kopia repo with around 100TB of data - everything was going well, it has not been touched for two weeks or so. I kicked of a small incremental backup today, which completed successfully: \* 0 hashing, 338337 hashed (878.7 GB), 28078694 cached (94.8 TB), uploaded 876.2 GB, estimated 22 TB (435.2%) 0s left Created snapshot with root k3ad2929159e95246105260ef08b97a3d818b4fc02c580e8816fefe8413946756 and ID 3bee2c042f4de4cb7088f9ef616462e8 in 36m59s However, after a while (maybe wh… > [@Help with corrupt repo](https://kopia.discourse.group/t/help-with-corrupt-repo/1247): > > Hi there, I’m wondering if I can get some help to recover a seemingly corrupt repo and also possibly some advice if the way I’m running maintenance led to the corruption. My backup script has this sequence of commands: kopia repository connect filesystem ... kopia snapshot create /path1 kopia snapshot create /path2 ... kopia snapshot expire --delete kopia maintenance run --full --force kopia repository sync-to ... On a recent run, one of the snapshot create steps failed with this error: ERR… Maybe Kopia just isn’t there yet … not going into depth how the filesystem-corruption can happen e.g. bitrot - just that it seems really hacky and off not being able to work with a system that’s built to fight such issues by design (one would think). Yet I’m still here … maybe these are just random errors … who knows. I’ll still follow a bit to see the experiences from others, and maybe try again … --- _[View the full topic](https://kopia.discourse.group/t/seeing-kopia-connect-crash-having-large-number-of-epochs/2189)._