git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: git error in tag ...: unterminated header

From
Junio C Hamano <gitster@pobox.com>
Date
Jun 25, 2015, 20:48 UTC
Message-ID
<xmqqa8vny0nf.fsf@gitster.dls.corp.google.com>
In-Reply-To
<20150625201309.5026A384E81@gemini.denx.de>
Wolfgang Denk <wd@denx.de> writes:
Show 7 quoted lines
> Hm... it seems I cannot even easily delte these tags:
>
> -> git tag -d LABEL_2006_03_12_0025
> Deleted tag 'LABEL_2006_03_12_0025' (was eb394f5)
> -> git fsck --full
> Checking object directories: 100% (256/256), done.
> Checking object directories: 100% (256/256), done.

This is expected. "git tag -d" only severed the linkage between tagname LABEL_2006_03_12_0025 and the tag object eb394f5; without repacking and pruning, the object eb394f5 is not removed from your object store.

> dangling tag eb394f56db3e05d00891d6dc36a00df0025cf255
>
> Now I also have this "dangling tag" thingy...

That also is expected. It now is "dangling" because it is not reachable from any of your refs.

"git repack -a -d -f" after "git tag -d && git tag -a" would make your repository fsck-clean, but I highly doubt that you would want to recreate your tags right now.

If I were you, I'd learn to ignore these 'unterminated header' errors for now and wait to see if Git folks decide that 4d0d8975 (Make sure fsck_commit_buffer() does not run out of the buffer, 2014-09-11) is overzealous and giving an error message where there is no error, in which case future versions of Git will not complain on these objects. It is not too late to do the re-tagging after they decide to keep the current behaviour.

Previous: Johannes SchindelinNext: Junio C Hamano
Message 18 of 19 in “git error in tag ...: unterminated header”
  1. Wolfgang DenkJun 25, 2015
  2. Junio C HamanoJun 25, 2015
  3. Wolfgang DenkJun 25, 2015
  4. Junio C HamanoJun 25, 2015
  5. Johannes SchindelinJun 25, 2015
  6. Junio C HamanoJun 25, 2015
  7. Junio C HamanoJun 25, 2015
  8. Johannes SchindelinJun 26, 2015
  9. Jeff KingJun 26, 2015
  10. Junio C HamanoJun 26, 2015
  11. Johannes SchindelinJun 27, 2015
  12. Junio C HamanoJun 27, 2015
  13. fsck: it is OK for a tag and a commit to lack the bodyJunio C Hamano, Jun 28, 2015
  14. Eric SunshineJun 28, 2015
  15. Johannes SchindelinJun 29, 2015
  16. Junio C HamanoJun 29, 2015
  17. Johannes SchindelinJun 29, 2015
  18. Junio C HamanoJun 25, 2015
  19. Junio C HamanoJun 25, 2015

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.