Commit Graph

546 Commits

Author SHA1 Message Date
Morris Jobke 48653d1a27
Merge pull request #16440 from marcelklehr/fix/objectstorage-put-contents
Fix File#putContents(string) on ObjectStorage
2019-07-17 22:38:41 +02:00
Marcel Klehr d46744e2f1 Fix File#putContents(string) on ObjectStorage
Signed-off-by: Marcel Klehr <mklehr@gmx.net>
2019-07-17 14:58:56 +02:00
Roeland Jago Douma 8ca2b31804
Do not keep searching for recent
If userA has a lot of recent files. But only shares 1 file with userB
(that has no files at all). We could keep searching until we run out of
recent files for userA.

Now assume the inactive userB has 20 incomming shares like that from
different users. getRecent then basically keeps consuming huge amounts
of resources and with each iteration the load on the DB increases
(because of the offset).

This makes sure we do not get more than 3 times the limit we search for
or more than 5 queries.

This means we might miss some recent entries but we should fix that
separatly. This is just to make sure the load on the DB stays sane.

Signed-off-by: Roeland Jago Douma <roeland@famdouma.nl>
2019-07-16 19:14:55 +02:00
Roeland Jago Douma e953205908
Use HTTP1.1 to read S3 objects
Some of the READs otherwise use HTTP/1.0 which is not always supported
by all backends. HTTP/1.1 is there since 1999 way longer than S3 so safe
to assume it is always there IMO.

Signed-off-by: Roeland Jago Douma <roeland@famdouma.nl>
2019-07-10 11:42:22 +02:00
John Molakvoæ 5a03189ce7
Mimetype list integrity check should not fail if it's changed (#15810)
Mimetype list integrity check should not fail if it's changed
2019-07-07 20:01:58 +02:00
Roeland Jago Douma 5312a07f55
Merge pull request #16107 from nextcloud/local-check-path
verify that paths are valid for recursive local move
2019-06-27 16:36:30 +02:00
Robin Appelman f39605dabc
verify that paths are valid for recursive local move
Signed-off-by: Robin Appelman <robin@icewind.nl>
2019-06-27 11:10:08 +02:00
Roeland Jago Douma 2f5810f26e
Merge pull request #15832 from nextcloud/bugfix/noid/fulltext-search-groupfolders
Fix full text search for groupfolders
2019-06-25 21:43:37 +02:00
Morris Jobke 9dac9710e4 Fix full text search for groupfolders
Signed-off-by: Morris Jobke <hey@morrisjobke.de>
2019-06-17 12:52:22 -01:00
Robin Appelman b4e2d8f8ad
handle storage exceptions when trying to set mtime
not all storage backends can handle setting the mtime and they might
not always handle that error correctly.

Signed-off-by: Robin Appelman <robin@icewind.nl>
2019-06-06 16:09:27 +02:00
Xheni Myrtaj d227f4d34c
Do not fail integrity check if mimetype list is changed
Signed-off-by: Xheni Myrtaj <myrtajxheni@gmail.com>
2019-05-29 22:43:23 +02:00
Roeland Jago Douma 58fe0b8032
Merge pull request #15776 from nextcloud/get-by-id-shares-last
search files by id in shared storages last
2019-05-29 21:56:15 +02:00
Robin Appelman 98dd526010
search files by id in shared storages last
this prevents unneeded recursion when the file is not in the share

Signed-off-by: Robin Appelman <robin@icewind.nl>
2019-05-28 20:03:45 +02:00
Julius Härtl 67dd4b018a
Check for free space on touch
Signed-off-by: Julius Härtl <jus@bitgrid.net>
2019-05-28 13:06:59 +02:00
Morris Jobke 96e892770d
Merge pull request #15715 from nextcloud/s3-bucket-name
Allow other code to use the s3 client from object store
2019-05-24 15:03:22 +02:00
Robin Appelman 834db3af2c
add method to get bucket name from s3 storage
Signed-off-by: Robin Appelman <robin@icewind.nl>
2019-05-24 14:02:03 +02:00
Robin Appelman 06a119242e
make ObjectStoreStorage::getURN public
Signed-off-by: Robin Appelman <robin@icewind.nl>
2019-05-22 12:17:33 +02:00
Robin Appelman 439b341e89
add method to get the used object store from the storage
Signed-off-by: Robin Appelman <robin@icewind.nl>
2019-05-21 17:18:00 +02:00
Robin Appelman 6a30c65a32
remove extra param in cache update call
param was leftover from earlier iterations of the "update event"

Signed-off-by: Robin Appelman <robin@icewind.nl>
2019-05-07 14:53:02 +02:00
Arthur Schiwon 009c2dfbd5
take return of mkdir into account, throw exception on failure
Signed-off-by: Arthur Schiwon <blizzz@arthur-schiwon.de>
2019-04-29 16:45:59 +02:00
Roeland Jago Douma dedb84acb0
Always use a temp file for swift writes
Apparently the if statement doesn't work in all cases (even if I could
not reproduce it). So for the time being we will just not directly
stream to swift.

Signed-off-by: Roeland Jago Douma <roeland@famdouma.nl>
2019-04-18 21:04:27 +02:00
Roeland Jago Douma 5d360bd16f
Harden appdata putcontent
If for whatever reason appdata got into a strange state this will at
least propegate up and not make it do boom the next run.

Signed-off-by: Roeland Jago Douma <roeland@famdouma.nl>
2019-04-08 13:22:00 +02:00
Vlastimil Pecinka 1c3abda280
Store user object later in the code
Signed-off-by: Vlastimil Pecinka <pecinka@email.cz>
2019-03-21 21:57:47 +01:00
Robin Appelman cc29f5d719
always allow moving mountpoints inside the same mountpoint
even if that mountpoint isn't normally a valid target for moving mounts into

Signed-off-by: Robin Appelman <robin@icewind.nl>
2019-03-19 15:56:18 +01:00
Roeland Jago Douma 6f88f28a15
Merge pull request #14689 from nextcloud/rename-moveable-mount-readonly-parent
allow renaming the root of a movable mount even if the parent folder …
2019-03-15 16:48:34 +01:00
Joas Schilling 17d5e47937
Remove spaming debug logs
Signed-off-by: Joas Schilling <coding@schilljs.com>
2019-03-15 12:13:44 +01:00
Robin Appelman 3ca1e90044
allow renaming the root of a movable mount even if the parent folder is readonly
Signed-off-by: Robin Appelman <robin@icewind.nl>
2019-03-14 15:27:25 +01:00
Robin Appelman 631ae17dce
handle long etags from dav external storage
we can only store etags up to 40 characters long in the database, so when we get an etag that's longer we simply hash it to bring down the length

Signed-off-by: Robin Appelman <robin@icewind.nl>
2019-03-14 14:46:39 +01:00
Morris Jobke 772303309d
Merge pull request #14425 from cowai/fix-multiple-incomplete-folders
Do not calculate folder size for parent that also needs proper scan, fixes #3524
2019-03-08 09:12:33 +01:00
Roeland Jago Douma 1b27e9578d
Merge pull request #14424 from nextcloud/fix/13554/swift_to_tmp
Use a tmp file for swift writes
2019-03-06 15:33:15 +01:00
Ari Selseng d16cfb519e Avoid calculating folder size for parent that needs scan.
Signed-off-by: Ari Selseng <ari@selseng.net>
2019-03-06 15:23:37 +01:00
Morris Jobke debd32b461
Merge pull request #14175 from skydiablo/master
Fix ObjectStorage expired token (v3)
2019-03-05 18:59:45 +01:00
Roeland Jago Douma 1153123b90
Use a tmp file for swift writes
Else this leads to a seekable stream error with chunked uploads

Signed-off-by: Roeland Jago Douma <roeland@famdouma.nl>
2019-03-04 13:11:59 +01:00
Julius Härtl c778032276
Properly handle null as ownerId if file system access is denied by access token
Signed-off-by: Julius Härtl <jus@bitgrid.net>
2019-03-01 20:57:01 +01:00
Joas Schilling 01b4db62fb
Add dispatcher events to User and Group objects
Signed-off-by: Joas Schilling <coding@schilljs.com>
2019-03-01 20:56:59 +01:00
Roeland Jago Douma df55781814
Merge pull request #14254 from sparrowjack63/master
Dont check Bucket Name in Nextcloud
2019-03-01 13:46:04 +01:00
S. Cat e3142a91b2 Bucket name - Alert in debug mode
Alert in debug mode only when trying to mount non compatible DNS bucket name (in order to not flood the logs)

Signed-off-by: Sébastien Cat <sebastien.cat@inra.fr>
2019-03-01 11:28:54 +01:00
S. Cat 381ae44d38 Correct code indent
Just correct code indent.

Signed-off-by: Sébastien Cat <sebastien.cat@inra.fr>
2019-03-01 11:28:54 +01:00
S. Cat 77473b0228 Warning on S3 not compatible DNS bucket name
The new created bucket should respect the DNS compatibility, nevertheless, Nextcloud should accept to mount "old created" buckets that does not respect DNS compatibility (Backward compatibility, or compatibility with CEPH).

Signed-off-by: Sébastien Cat <sebastien.cat@inra.fr>
2019-03-01 11:28:54 +01:00
Sebastien Cat 59c3d8d85f Dont check Bucket Name in Nextcloud
Signed-off-by: Sebastien Cat <sparrow.jack63@gmail.com>
Signed-off-by: Sébastien Cat <sebastien.cat@inra.fr>
2019-03-01 11:28:54 +01:00
Robin Appelman 8fc47c6f00
add option to get raw size (without submounts) from fileinfo
Signed-off-by: Robin Appelman <robin@icewind.nl>
2019-02-27 15:35:44 +01:00
Morris Jobke 50d7f87eab
Merge pull request #14269 from nextcloud/cache-update-event
add event for cache update
2019-02-19 11:07:59 +01:00
Robin Appelman 092f78d7a4
add event for cache update
Signed-off-by: Robin Appelman <robin@icewind.nl>
2019-02-18 16:50:41 +01:00
Roeland Jago Douma f6f002e2ad
Merge pull request #14210 from nextcloud/fix/14192/fix_empty_uploads
Fix empty file uploads to S3 (and other streaming storages)
2019-02-18 15:39:17 +01:00
Roeland Jago Douma 104616c1da
Use CountWrapper instead of almost identical CountReadStream
Signed-off-by: Roeland Jago Douma <roeland@famdouma.nl>
2019-02-15 13:24:58 +01:00
Roeland Jago Douma 61d66d7954
Fix empty file uploads to S3
Signed-off-by: Roeland Jago Douma <roeland@famdouma.nl>
2019-02-15 08:41:39 +01:00
Roeland Jago Douma 62ddb124a1
Filter out versions and trashbin
Signed-off-by: Roeland Jago Douma <roeland@famdouma.nl>
2019-02-13 21:27:25 +01:00
Roeland Jago Douma cb67c6aa00
Loop for the recent search
Else it might happen that less entries than requested are returned

Signed-off-by: Roeland Jago Douma <roeland@famdouma.nl>
2019-02-13 21:22:16 +01:00
Roeland Jago Douma 4c7424d38b
We need to initialize the storage else the jailed path is always null
Signed-off-by: Roeland Jago Douma <roeland@famdouma.nl>
2019-02-13 21:01:17 +01:00
Volker d06c25b9c8
Update SwiftFactory.php
Auth-Service (in explizit v3) will recheck the cached-token and will end in an "token expired exception".
2019-02-13 12:26:27 +01:00