mod_dav: optionally set mtime against x-oc-mtime header #556

leo9800 · 2025-09-04T04:43:03Z

WebDAV RFCs did not specify a standard approach to set modification time for files or directory on the server, this feature enables better presevation of file metadata when uploading.

as a compensation, OwnCloud (now known as NextCloud) proposed a X-OC-Mtime header which contains UNIX timestamp of mtime with file-modifying methods (e.g. PUT) which implements the feature above as a extension to the standard.

this patch allows apache to optionally honor the X-OC-Mtime header and set mtime accordingly if DavHonorMtimeHeader On is set.

notroj

This is an interesting feature. Are you aware of any standardisation effort for that header?

modules/dav/main/mod_dav.c

notroj · 2025-09-08T10:05:56Z

modules/dav/fs/repos.c

+    status = apr_file_mtime_set(resource->info->pathname, mtime, pool);
+
+    if (status != APR_SUCCESS) {
+        return dav_new_error(pool, HTTP_BAD_REQUEST, 0, status, "Could not set mtime.");


If we get this far this is likely not a client error so this should be mapped to something appropriate rather than a 400. Or default to 500?

actually i was a bit unsure whether 4xx or 5xx should be returned in such case. despite most invalid patterns has already being sanitized in dav_parse_mtime, including:

non-digits, e.g. nonsense bytes

negatives, e.g. -123123123

non-decimal, e.g. 0xa0b1

empty strings

overflowing

while apr_file_mtime_set may still fail, in most cases it is clear that a 500 should be returned, e.g. kernel VFS failed to respond utime syscall. but i was wondering if there is any possibility that an invalid user input fails apr_file_mtime_set where 400 should be returned.

as of now I would return a 500 in such cases instead of 400.

leo9800 · 2025-09-09T01:41:14Z

This is an interesting feature. Are you aware of any standardisation effort for that header?

seems there is not much such efforts (i.e. RFC drafts) from what i know.

it is an approach OwnCloud (predecessor of NextCloud) adopts to enable mtime presevation when uploading file via WebDAV as RFCs of WebDAV does not define a standard way to do so.

this feature is also implemented by rclone's rclone serve webdav WebDAV server command: https://github.com/rclone/rclone/blob/3f0e9f5fca8080f87094341f8ac5027091cc7601/cmd/serve/webdav/webdav.go#L339

given that this is a non-standard, i would implement it but keep it disabled by default to avoid any non-standard breaking change. users who prefer to use this feature should have it opted in manually with DAVHonorMtimeHeader On.

notroj

Thanks for the update, looking good. One minor style nit:

} else {

should be written as

}
else {

modules/dav/fs/repos.c

modules/dav/main/mod_dav.c

notroj · 2025-09-22T15:21:42Z

modules/dav/main/mod_dav.c

 }

+/**
+ * @return  1 if valid x-oc-mtime,


I would prefer this with an apr_status_t return value, though it's a minor question of style

i am not sure if apr_status_t is proper for this case personally.

since 0 is defined as APR_SUCCESS, while for the returning value of this function both 1 and 0 are considered success, and -1 stands for a failure.

actually i was imitating static int dav_parse_range(request_rec *r, apr_off_t *range_start, apr_off_t *range_end) in mod_dav.c

modules/dav/main/mod_dav.c

ref: apache#556 (review) Signed-off-by: Leo <[email protected]>

leo9800 · 2025-09-23T09:58:54Z

@notroj

i am still a bit confused on how should APLOGNO() being handled with APLOG_DEBUG leveled logs.

there are 3 possible handlings:

ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO() "PLACEHOLDER MSG");: do not assign tags, let apache maintainers to invoke docs/log-message-tags/update-log-msg-tags
ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(01234) "PLACEHOLDER MSG");: assign tags, by invoking docs/log-message-tags/update-log-msg-tags
ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, "PLACEHOLDER MSG");: do not include APLOGNO() at all

the comment

as above don't use APLOGNO() for DEBUG-level logging

suggests approach 3, but by invoking git grep -F APLOG_DEBUG on the repo huge amount of modules included APLOGNO(), which suggests approach 1 or 2.

modules/aaa/mod_auth_digest.c:        ap_log_error(APLOG_MARK, APLOG_DEBUG, 0, NULL, APLOGNO(01757)
modules/aaa/mod_auth_digest.c:    ap_log_error(APLOG_MARK, APLOG_DEBUG, 0, cmd->server, APLOGNO(01763)
modules/aaa/mod_auth_digest.c:        ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(01764)
modules/aaa/mod_auth_digest.c:        ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(01765)
modules/aaa/mod_auth_digest.c:    ap_log_error(APLOG_MARK, APLOG_DEBUG, 0, s, APLOGNO(01768)
modules/aaa/mod_auth_form.c:        ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(02982)
modules/aaa/mod_auth_form.c:        ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(02983)
modules/aaa/mod_auth_form.c:        ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(01812)
modules/aaa/mod_authn_socache.c:        ap_log_rerror(APLOG_MARK, APLOG_DEBUG, rv, r, APLOGNO(01679)
modules/aaa/mod_authn_socache.c:        ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(01681)
modules/aaa/mod_authn_socache.c:        ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(01684)
modules/aaa/mod_authn_socache.c:        ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(01685)
modules/aaa/mod_authn_socache.c:        ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(01687)
modules/aaa/mod_authn_socache.c:        ap_log_rerror(APLOG_MARK, APLOG_DEBUG, 0, r, APLOGNO(01688)
# truncated git grep output ...

notroj · 2025-09-30T10:21:29Z

(3). For DEBUG-level or TRACE-level , omit the APLOGNO() entirely. This was the consensus on the dev@ list, possibly isn't documented anywhere.

leo9800 · 2025-09-30T12:54:57Z

@notroj gotcha

leo9800 · 2025-09-30T13:10:25Z

should i also rebase this patch to current trunk?

i noticed a merge conflict related to log-number, (APLOGNO) probably caused by another patch submitted later than this introducing some new log messages.

notroj · 2025-10-02T12:28:35Z

Don't worry about APLOGNO() conflicts, leave that to a committer to fix up. I did a CI run - one issue here: https://github.com/notroj/httpd/actions/runs/18156145629/job/51676367710
move the apr_size_t to the top of the function.

leo9800 · 2025-10-02T13:41:38Z

@notroj gotcha

seems we need to ensure conformity with C89/C90

notroj · 2025-10-02T15:19:57Z

Some more failures now https://github.com/notroj/httpd/actions/runs/18197214275/job/51806547259#logs

and probably best to rebase onto the trunk again because the trunk tests are failing without 60ad836

ref: apache#556 (review) Signed-off-by: Leo <[email protected]>

…av_hooks_repository` modules/dav/fs/repos.c: implement `dav_fs_set_mtime` Signed-off-by: Leo <[email protected]>

Signed-off-by: Leo <[email protected]>

…mtime when `conf->honor_mtime_header == DAV_ENABLED_ON`, check `set_mtime` callback is available before invoking, log debug-level messages docs/log-message-tags: bumped with `docs/log-message-tags/update-log-msg-tags` Signed-off-by: Leo <[email protected]>

…e_set` fails, emit error-level log message also docs/log-message-tags: bumped with `docs/log-message-tags/update-log-msg-tags` Signed-off-by: Leo <[email protected]>

Signed-off-by: Leo <[email protected]>

…y beginning of `dav_method_put()`, this avoids access of uninitialized structure Signed-off-by: Leo <[email protected]>

Signed-off-by: Leo <[email protected]>

ref: apache#556 (review) Signed-off-by: Leo <[email protected]>

…invocation is not required Signed-off-by: Leo <[email protected]>

…ssary terminologies Signed-off-by: Leo <[email protected]>

Signed-off-by: Leo <[email protected]>

…for debug and trace level logs Signed-off-by: Leo <[email protected]>

…-msg-tags` invocation Signed-off-by: Leo <[email protected]>

…cation times for directory (MKCOL) Signed-off-by: Leo <[email protected]>

Signed-off-by: Leo <[email protected]>

- remove unnecessary variable declarations: `mtime_aware`, `err3` - always initialize `mtime_ret` to `0` - simplify mtime-setting procedures Signed-off-by: Leo <[email protected]>

…OGNO()` after rebase Signed-off-by: Leo <[email protected]>

notroj · 2025-10-06T10:43:33Z

https://github.com/notroj/httpd/actions/runs/18216313264

🎉 I'll try to get it merged this week, thanks for your persistence!

leo9800 · 2025-10-07T02:16:52Z

@notroj thanks for reviewing, cheers🎉

also, kindly asking, could you please review another patch initiated by me at #558? (probably try with CI builds, etc)

leo9800 marked this pull request as ready for review September 6, 2025 19:56

notroj reviewed Sep 8, 2025

View reviewed changes

leo9800 requested a review from notroj September 9, 2025 02:32

notroj suggested changes Sep 22, 2025

View reviewed changes

leo9800 added a commit to leo9800/httpd that referenced this pull request Sep 23, 2025

modules/dav/main/mod_dav.c: code style fixes

1269571

ref: apache#556 (review) Signed-off-by: Leo <[email protected]>

leo9800 requested a review from notroj October 2, 2025 14:49

leo9800 added a commit to leo9800/httpd that referenced this pull request Oct 2, 2025

modules/dav/main/mod_dav.c: code style fixes

336ee63

ref: apache#556 (review) Signed-off-by: Leo <[email protected]>

leo9800 force-pushed the x-oc-mtime branch from 0a82255 to b72db41 Compare October 2, 2025 16:51

leo9800 added 14 commits October 3, 2025 00:53

modules/dav/main/mod_dav.h: define callback set_mtime for `struct d…

de7be9c

…av_hooks_repository` modules/dav/fs/repos.c: implement `dav_fs_set_mtime` Signed-off-by: Leo <[email protected]>

modules/dav/main/mod_dav.c: add new config flag DavHonorMtimeHeader

e77869f

Signed-off-by: Leo <[email protected]>

modules/dav/main/mod_dav.c: implement dav_parse_range

86e42ae

Signed-off-by: Leo <[email protected]>

modules/dav/main/mod_dav.c: parse x-oc-mtime for PUT requests

177801f

Signed-off-by: Leo <[email protected]>

modules/dav/fs/repos.c: return 500 instead of 400 upon `apr_file_mtim…

8217af3

…e_set` fails, emit error-level log message also docs/log-message-tags: bumped with `docs/log-message-tags/update-log-msg-tags` Signed-off-by: Leo <[email protected]>

modules/dav/main/mod_dav.c: fix type on request method in error message

32a4cce

Signed-off-by: Leo <[email protected]>

modules/dav/main/mod_dav.c: bugfix: retrieve module config at the ver…

76ee8e0

…y beginning of `dav_method_put()`, this avoids access of uninitialized structure Signed-off-by: Leo <[email protected]>

modules/dav/main/mod_dav.c: parse x-oc-mtime for MKCOL requests

16ec946

Signed-off-by: Leo <[email protected]>

modules/dav/main/mod_dav.c: code style fixes

18c79d2

ref: apache#556 (review) Signed-off-by: Leo <[email protected]>

modules/dav/main/mod_dav.c: zero-ing errno before apr_strtoi64() …

b776d4d

…invocation is not required Signed-off-by: Leo <[email protected]>

modules/dav/main/mod_dav.c: use proper language in logs, avoid unnece…

44959b9

…ssary terminologies Signed-off-by: Leo <[email protected]>

modules/dav/fs/repo.c: avoid unnecessary apr_strerror invocation

a14efe0

Signed-off-by: Leo <[email protected]>

modules/dav/fs/repos.c, modules/dav/main/mod_dav.c: remove APLOGNO() …

de608fd

…for debug and trace level logs Signed-off-by: Leo <[email protected]>

leo9800 added 5 commits October 3, 2025 00:53

modules/dav/fs/repo.c, docs/log-message-tags/next-number: `update-log…

ca3b037

…-msg-tags` invocation Signed-off-by: Leo <[email protected]>

modules/dav/main/mod_dav.c: rephrase log messages when setting modifi…

f2d3df8

…cation times for directory (MKCOL) Signed-off-by: Leo <[email protected]>

modules/dav/main/mod_dav.c: dav_parse_mtime(): ensure C89 conformity

87ce719

Signed-off-by: Leo <[email protected]>

modules/dav/main/mod_dav.c: compile errors fixes

8170906

- remove unnecessary variable declarations: `mtime_aware`, `err3` - always initialize `mtime_ret` to `0` - simplify mtime-setting procedures Signed-off-by: Leo <[email protected]>

modules/dav/fs/repos.c, docs/log-message-tags/next-number: reset `APL…

f4128dd

…OGNO()` after rebase Signed-off-by: Leo <[email protected]>

leo9800 force-pushed the x-oc-mtime branch from b72db41 to f4128dd Compare October 2, 2025 16:53

mod_dav: optionally set mtime against x-oc-mtime header #556

Are you sure you want to change the base?

mod_dav: optionally set mtime against x-oc-mtime header #556

Conversation

leo9800 commented Sep 4, 2025

Uh oh!

notroj left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

notroj Sep 8, 2025

Choose a reason for hiding this comment

Uh oh!

leo9800 Sep 9, 2025

Choose a reason for hiding this comment

Uh oh!

leo9800 commented Sep 9, 2025

Uh oh!

notroj left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

notroj Sep 22, 2025

Choose a reason for hiding this comment

Uh oh!

leo9800 Sep 23, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

leo9800 commented Sep 23, 2025

Uh oh!

notroj commented Sep 30, 2025

Uh oh!

leo9800 commented Sep 30, 2025

Uh oh!

leo9800 commented Sep 30, 2025

Uh oh!

notroj commented Oct 2, 2025

Uh oh!

leo9800 commented Oct 2, 2025

Uh oh!

notroj commented Oct 2, 2025

Uh oh!

notroj commented Oct 6, 2025

Uh oh!

leo9800 commented Oct 7, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

leo9800 Sep 23, 2025 •

edited

Loading