[...] does not exist when running `prune`

[teidesu]

not sure how i can make an mcve for this nor debug any further, but i randomly started getting the above error when running rustic prune --instant-delete

on debug log level no additional output got printed, a portion of logs from TRACE level:

$ rustic prune --instant-delete --log-level TRACE
[...a bunch of similar log lines with different Index/Snapshot/Pack ids...]
[TRACE] (34) rustic_core::backend::cache: [/build/rustic-0.10.1-vendor/rustic_core-0.9.0/src/backend/cache.rs:469] cache reading tpe: Pack, id: 0698cdf5, offset: 1085301
[TRACE] (34) rustic_core::backend::cache: [/build/rustic-0.10.1-vendor/rustic_core-0.9.0/src/backend/cache.rs:520] cache hit!
[TRACE] (34) rustic_core::backend::cache: [/build/rustic-0.10.1-vendor/rustic_core-0.9.0/src/backend/cache.rs:469] cache reading tpe: Pack, id: 169877d1, offset: 0
[TRACE] (34) rustic_core::backend::cache: [/build/rustic-0.10.1-vendor/rustic_core-0.9.0/src/backend/cache.rs:520] cache hit!
[TRACE] (34) rustic_core::backend::cache: [/build/rustic-0.10.1-vendor/rustic_core-0.9.0/src/backend/cache.rs:469] cache reading tpe: Pack, id: 4f8c8934, offset: 0
[TRACE] (34) rustic_core::backend::cache: [/build/rustic-0.10.1-vendor/rustic_core-0.9.0/src/backend/cache.rs:520] cache hit!
[TRACE] (1) rustic_backend::local: [/build/rustic-0.10.1-vendor/rustic_backend-0.5.4/src/local.rs:247] listing tpe: Pack
error: `rustic_core` experienced an error related to `internal operations`.

Message:
Pack `65dcc0d3` does not exist.

We believe this is a bug, please report it by opening an issue at:
https://github.com/rustic-rs/rustic_core/issues/new

If you can, please attach an anonymized debug log to the issue.

Thank you for helping us improve rustic!


Some additional details ...

Backtrace:
   0: rustic_core::commands::prune::PrunePlan::check_existing_packs::{{closure}}
   1: rustic_rs::commands::prune::PruneCmd::inner_run
   2: rustic_rs::repository::AllRepositoryOptions::run_with_progress
   3: <rustic_rs::commands::prune::PruneCmd as abscissa_core::runnable::Runnable>::run
   4: <rustic_rs::commands::EntryPoint as abscissa_core::runnable::Runnable>::run

(the problematic Pack doesn't appear anywhere in the TRACE logs)

im running rustic 0.10.1 on NixOS 25.11, and i started getting this issue very soon after updating, so might be a regression?

---

running rustic check yields:

[INFO] using no config file, none of these exist: /home/teidesu/.config/rustic/rustic.toml, /etc/rustic/rustic.toml, ./rustic.toml
[INFO] repository local:[...]: password is correct.
[INFO] using cache at /home/teidesu/.cache/rustic/7a21e2678196672aa76bd4639c4fac3876fe1f7caf258ac6e980f3d1f6391d71
[00:00:00] getting snapshots...           ████████████████████████████████████████        125/125                                                                                                               [00:00:00] checking Snapshot in cache...  ████████████████████████████████████████  71.50 KiB/71.50 KiB  95.57 MiB/s  (ETA -)                                                                                   [00:00:00] checking Index in cache...     ████████████████████████████████████████  14.47 MiB/14.47 MiB  4.54 GiB/s   (ETA -)                                                                                   [00:00:00] reading index...               ████████████████████████████████████████        431/431                                                                                                               [00:00:00] listing packs...               ⠁                                                                                                                                                                     [ERROR] pack 2bf44089 is referenced by the index but not present! To repair: 'rustic repair index'.
[ERROR] pack 65dcc0d3 is referenced by the index but not present! To repair: 'rustic repair index'.
[00:00:00] listing packs...                                                                                                                                                                                     [00:00:00] cleaning up packs from cache...                                                                                                                                                                      [00:00:00] checking packs in cache...     ████████████████████████████████████████ 161.48 MiB/161.48 MiB 3.99 GiB/s   (ETA -)                                                                                   [00:00:01] checking trees...              ████████████████████████████████████████        125/125                                                                                                               error: `rustic_core` experienced an error related to `the repository`.

Message:
check found errors!


Some additional details ...

Backtrace:
disabled backtrace (set 'RUST_BACKTRACE="1"' environment variable to enable)

running rustic repare index removes the missing packs from the index, but rustic prune still doesn't work with a similar error

Message:
Blob ID `001de8e9` is missing in index files.

and running rustic check now results in a lot of

[ERROR] file "some.db" blob e4bb11c3 is missing in index
[ERROR] file "some.db" blob 3714c965 is missing in index
[ERROR] file "some.db" blob 7e3774a7 is missing in index
[ERROR] file "some.db" blob 402a3d73 is missing in index
[...repeated like 50 more times with different blob ids but the same file...]

and sure enough 001de8e9 does appear there (multiple times even)

i'm not sure how to fix missing blobs since there isn't rustic repair blobs. removing cache files doesn't help.

[bombela]

I think I have a similar issue?

I have about 12snapshots in there less than 400GB of data. I deleted a snapshot that was too big (because glob excludes did not apply because in the config they were absolute path, but I ran rustic with a relative backup path). Then I ran a prune.

rustic prune --repack-cacheable-only true
[INFO] using config [...]
[INFO] repository opendal:: password is correct.
[INFO] using cache at /var/cache/rustic/[...]
[00:00:52] reading index...               ████████████████████████████████████████        105/105
[00:00:10] reading snapshots...           ████████████████████████████████████████         19/19
[00:00:02] finding used blobs...          ████████████████████████████████████████         19/19
[00:36:35] getting packs from repository...
error: `rustic_core` experienced an error related to `internal operations`.

Message:
Pack size `0` of id `3f87bb7b` does not match the expected size `53705837` in the index file.

I am running another one with --log-level TRACE and RUST_BACKTRACE=full.

[bombela]

--log-level TRACE is useless. It doesn't even tell you what it is retrieving.

[00:05:18] getting packs from repository... ⠖
[TRACE] (1) reqwest::retry: [/home/runner/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/reqwest-0.12.24/src/retry.rs:231] shouldn't retry!
[TRACE] (7) hyper_util::client::legacy::pool: [/home/runner/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/hyper-util-0.1.17/src/client/legacy/pool.rs:348] put; add idle connection for ("https", www.googleapis.com)
[DEBUG] (7) hyper_util::client::legacy::pool: pooling idle connection for ("https", www.googleapis.com)
[TRACE] (1) hyper_util::client::legacy::pool: [/home/runner/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/hyper-util-0.1.17/src/client/legacy/pool.rs:653] take? ("https", www.googleapis.com): expiration = Some(90s)
[DEBUG] (1) hyper_util::client::legacy::pool: reuse idle connection for ("https", www.googleapis.com)
[00:05:19] getting packs from repository... ⠐

Furthermore, it takes almost 40 minutes for every test. I am not sure why it is so slow (uploading backups is many times faster, but it could be here some google drive shenanigan for what I know).

[teidesu]

it somehow fixed itself after a few days for me btw 🥴

[bombela]

Good to know. Still broken for me so far. Not sure why the backtrace is useless.

[00:34:07] getting packs from repository... ⠉
[TRACE] (1) reqwest::retry: [/home/runner/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/reqwest-0.12.24/src/retry.rs:231] shouldn't retry!
[TRACE] (7) hyper_util::client::legacy::pool: [/home/runner/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/hyper-util-0.1.17/src/client/legacy/pool.rs:348] put; add idle connection for ("https", www.googleapis.com)
[DEBUG] (7) hyper_util::client::legacy::pool: pooling idle connection for ("https", www.googleapis.com)
[DEBUG] (1) opendal::services: service=gdrive name= path=data/ listed=8870: list finished
[00:34:07] getting packs from repository...
error: `rustic_core` experienced an error related to `internal operations`.

Message:
Pack size `0` of id `3f87bb7b` does not match the expected size `53705837` in the index file.

We believe this is a bug, please report it by opening an issue at:
https://github.com/rustic-rs/rustic_core/issues/new

If you can, please attach an anonymized debug log to the issue.

Thank you for helping us improve rustic!


Some additional details ...

Backtrace:
   0: <unknown>
   1: <unknown>
   2: <unknown>
   3: <unknown>
   4: <unknown>
   5: <unknown>
   6: <unknown>
   7: <unknown>
   8: <unknown>
   9: <unknown>
  10: <unknown>
  11: __libc_start_main
  12: <unknown>

[bombela]

And here is part of the output of rustic repair index. This does not bode well.

[bombela]

Pretty much all messages say something similar to

[WARN] error reading pack 496d656b (-> removing from index): Error: Partially reading file `data/49/496d656b20fb836b8a1c2010ea7296f5c10122ed92b1f9cc2847b83105756dcc` failed in the backend. Please check if the given path is correct. (kin
d: related to the backend): caused by: Unexpected (persistent) at , context: { expect: 18446744069414585996, actual: 53635327 } => reader got too little data

If they are byte amount, 18446744069414585996 B = 16 EiB, and 53635327 B ≈ 51.15063381 MiB. I doubt that my pack files are 16 EiB large.

[bombela]

I cancelled the repair job, and started a backup (snapshot), and got this warning:

[WARN] ignoring error when loading parent tree 1864361b: Error: Tree ID `1864361b` not found in index (kind: related to internal operations)

Then the backup still started, but it seemed to be re-uploading everything and doubling the size of the 400GB repository.

[aawsome]

@bombela could it be that you are affected by #1619?

[bombela]

Thanks for the link, this has to be this indeed. And the more I tried to repair de repo, the worst is became.

…

On Thu, Dec 18, 2025, 10:44 aawsome ***@***.***> wrote: *aawsome* left a comment (rustic-rs/rustic#1613) <#1613 (comment)> @bombela <https://github.com/bombela> could it be that you are affected by #1619 <#1619>? — Reply to this email directly, view it on GitHub <#1613 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AABUZDBOS6O2JBRKUSL45W34CLYYLAVCNFSM6AAAAACNZZWK7WVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTMNZRG4YDEOBXGQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[teidesu]

@aawsome for me it most likely wasn't, i use rustic with a local filesystem. though i can't really repro this anymore either, it seems to have fixed itself. perhaps this should be closed...