2024-06-27T07:15:28.686Z | <Xiubo Li> Hi Erich, please upload it via `ceph-post-file` |
2024-06-27T10:59:49.486Z | <p_the_b> Hi! I'm looking for an Object Storage (filesystem) to de-embed descriptive metadata on an object level (like Dublin Core fields) - for long-term preservation. |
2024-06-27T11:02:24.189Z | <p_the_b> I've searched the web up and down - and keep running in circles: Everything points to S3 - but S3 has harsh limitations on metadata (lowercase, 2k total, 10 tags, 128 chars each, etc) |
2024-06-27T11:03:09.897Z | <p_the_b> I'm grateful for any input on how to really use the metadata awesomeness of Object Storages :) |
2024-06-27T11:30:42.550Z | <Dhairya Parmar> hey @Venky Shankar there? |
2024-06-27T11:30:56.890Z | <Venky Shankar> yes |
2024-06-27T11:32:31.733Z | <Dhairya Parmar> (re: journal reset for tracker 61009 issue - preallocated inodes), does this guide look good to go: <https://termbin.com/swzc> |
2024-06-27T11:33:30.381Z | <Venky Shankar> yeh |
2024-06-27T11:33:35.764Z | <Venky Shankar> add a scrub to be run on the fs |
2024-06-27T11:33:56.184Z | <Dhairya Parmar> post journal reset and before bringing back the fs online? |
2024-06-27T11:34:08.832Z | <Venky Shankar> scrub requires online mds |
2024-06-27T11:34:15.878Z | <Dhairya Parmar> okay |
2024-06-27T11:41:02.543Z | <Dhairya Parmar> okay 😅 |
2024-06-27T12:58:20.882Z | <Dhairya Parmar> @Venky Shankar @Xiubo Li we don't test `Client::ll_write_block` anywhere? Do you folks have any idea about this io code path? it accepts `snapseq` and file layout which are i haven't seen in other i/o code paths. |
2024-06-27T12:59:33.586Z | <Dhairya Parmar> @Venky Shankar @Xiubo Li we don't test `Client::ll_write_block` anywhere? Do you folks have any idea about this io code path? it accepts `snapseq` and file layout which i haven't seen in other i/o code paths. |
2024-06-27T13:02:25.083Z | <Xiubo Li> It seems no test for this. |
2024-06-27T13:03:25.467Z | <Venky Shankar> what's ll_Write_block? |
2024-06-27T13:03:48.830Z | <Venky Shankar> oh it accepts a blockid |
2024-06-27T13:03:51.892Z | <Xiubo Li> safe write |
2024-06-27T13:04:12.302Z | <Xiubo Li> which will be blocked until successfully write to OSD |
2024-06-27T13:05:33.078Z | <Venky Shankar> how is that diff from a normal write? |
2024-06-27T13:05:41.468Z | <Venky Shankar> oh |
2024-06-27T13:05:55.018Z | <Venky Shankar> so to bypass buffered write? |
2024-06-27T13:06:01.235Z | <Venky Shankar> if its has caps? |
2024-06-27T13:06:20.031Z | <Xiubo Li> yeah |
2024-06-27T13:16:43.070Z | <Xiubo Li> @Dhairya Parmar This is normal with the other APIs, it will accept the layout to calculate the object extents and accept the snapseq to handle the snapshot |
2024-06-27T13:16:56.363Z | <Xiubo Li> @Dhairya Parmar This is normal and almost the same with the other APIs, it will accept the layout to calculate the object extents and accept the snapseq to handle the snapshot |
2024-06-27T13:34:28.939Z | <Dhairya Parmar> @Xiubo Li this looks to be more advanced since it allow controlling snapshot |
2024-06-27T13:35:08.951Z | <Dhairya Parmar> the thing is in all the sync/sync io patched i've pushed, this was never considered(i dont know this just hid from me) |
2024-06-27T13:41:10.443Z | <Venky Shankar> Hey devs RFR - <https://github.com/ceph/ceph/pull/58090> (cc @Patrick Donnelly) |
2024-06-27T13:44:24.695Z | <Dhairya Parmar> <https://github.com/ceph/ceph/commit/dad91b5ea7a50e752050a3dc9bd800337932aeb0>
im curious if this needs code coverage from our end |
2024-06-27T14:01:15.866Z | <Erich Weiler> Great! I uploaded the file with:
`ceph-post-file -d 'Tracker Bug 65607' dumps.tar.gz`
And after it was done I saw this:
`/usr/bin/ceph-post-file: copy the upload id below to share with a dev:`
`ceph-post-file: fb9a96f9-5f6d-46b4-b1fa-8580928b2241` |
2024-06-27T14:01:34.336Z | <Erich Weiler> Great! I uploaded the file with:
`ceph-post-file -d 'Tracker Bug 65607' dumps.tar.gz`
And after it was done I saw this:
```/usr/bin/ceph-post-file: copy the upload id below to share with a dev:
ceph-post-file: fb9a96f9-5f6d-46b4-b1fa-8580928b2241``` |
2024-06-27T14:03:22.625Z | <Erich Weiler> Great! @Xiubo Li I uploaded the file with:
`ceph-post-file -d 'Tracker Bug 65607' dumps.tar.gz`
And after it was done I saw this:
```/usr/bin/ceph-post-file: copy the upload id below to share with a dev:
ceph-post-file: fb9a96f9-5f6d-46b4-b1fa-8580928b2241``` |
2024-06-27T15:29:14.998Z | <Patrick Donnelly> looking now |
2024-06-27T15:31:05.006Z | <Patrick Donnelly> approved! |
2024-06-27T16:08:46.629Z | <Patrick Donnelly> ```+ git push -u origin wip-66585-squid
ssh: connect to host [github.com](http://github.com) port 22: Connection timed out
fatal: Could not read from remote repository.
Please make sure you have the correct access rights
and the repository exists.``` |
2024-06-27T16:08:50.307Z | <Patrick Donnelly> ugh this is so aggravating |
2024-06-27T16:29:45.767Z | <Venky Shankar> Is this from vossi01? |
2024-06-27T16:29:52.353Z | <Venky Shankar> I ran into this today morning my ime |
2024-06-27T16:29:53.612Z | <Venky Shankar> time* |