dnbd3.git/src, branch v2.1Distributed Network Block Device 3 --rewrite for Kernel 3.xhttps://git.openslx.org/dnbd3.git/atom/src?h=v2.12016-02-01T11:41:11+00:00[SERVER] BREAKING: Get rid of pseudo case-insensitivity2016-02-01T11:41:11+00:00Simon Rettberg2016-02-01T11:41:11+00:00urn:sha1:81c9da3c9d55fab13ab9a912cade320015d01613
This was a wrong decision made long time ago, and it's broken in
certain scenarios (eg. two servers serving from same NFS mount).
Also it's of limited use anyways since it only supportes ASCII and
would ignore umlauts, so blöd and BLÖD would still be considered
two different images.
So if you relied on this "feature" in any way, be careful when
updating.
[SERVER] Fix a lot of (mostly harmless) data races2016-01-11T11:09:23+00:00Simon Rettberg2016-01-11T11:09:23+00:00urn:sha1:d9c2a6cf943ca08f31f61a3fada940f77e3a03d3[KERNEL/CLIENT] Several minor tweaks and changes2016-01-11T10:59:03+00:00Simon Rettberg2016-01-11T10:59:03+00:00urn:sha1:aedf8c534195772f60e7c6409ae045fa20415aef
- Disable panic timeout that reports errors to block layer by default
- Get rid of "is_server" used by old proxy mode, introduce
"honor_server_alts" that just controls whether we use alts provided by
the server
- Allow switching servers when we only have 3 RTT measurements
- If using alts provided by server, don't ignore those that come from
other servers than the initial one
[FUSE] Handle SIGINT/SIGTERM to abort pending reads2015-12-17T15:33:22+00:00Simon Rettberg2015-12-17T15:33:22+00:00urn:sha1:b47ba01818f2ea2ebd1cf1bf5e3e91e37c5f88dd
Before, we would wait endlessly if there is a pending
read request that doesn't get answered (e.g. because
the server went down. That means you couldn't exit
the client in that case. Now we use a signal handler
to set a flag which causes the read to bail out and
return EIO.
[SERVER] Performance: Optimized some functions (gprof)2015-12-17T14:58:06+00:00Simon Rettberg2015-12-17T14:58:06+00:00urn:sha1:1460b746b6f5482ce1c56a30af232e824e316e56
A run with gprof revealed that background replication is a huge CPU hog.
The block selection was very slow and has been improved a lot.
Minor improvements were made to other functions that scan the cache map
of an image and are thus relatively slow.
[SERVER] image_markComplete now handles locking so we remove() unlocked2015-12-17T13:20:12+00:00Simon Rettberg2015-12-17T13:20:12+00:00urn:sha1:3de54fac10b3dbdcfc3573e175078c7104837ef3[SERVER] Cancel scanning of image dir if _shutdown is set2015-12-16T17:20:55+00:00Simon Rettberg2015-12-16T17:20:55+00:00urn:sha1:2bc1381256c860b6deb3044d3263cea625118793
So you can cancel image loading on startup via Ctrl-C
[SERVER] Fix image_calcBlockCrc32: Use 64bit for fileSize param2015-12-16T17:09:04+00:00Simon Rettberg2015-12-16T17:09:04+00:00urn:sha1:b5851655ab5f1cff84221e69fb6a74326b80bc22
Using uint32_t for fileSize is not too clever :(
[SERVER] More fine grained locking for RPC; better error logging2015-12-16T16:28:52+00:00Simon Rettberg2015-12-16T16:28:52+00:00urn:sha1:f49b63e11de50e72f85f8c6688da36d89bf17b87[FUSE] Fix forking mode (not passing -f) by not spawning threads before entering fuse_main2015-12-16T10:13:39+00:00Simon Rettberg2015-12-16T10:13:39+00:00urn:sha1:379436433bb4c9257a0af96025034065da623a0f