diff options
author | 2015-02-27 11:16:44 -0400 | |
---|---|---|
committer | 2015-02-27 11:16:44 -0400 | |
commit | f5ccb00d28d16ceb43b667e1f2d4f9759d22cba9 (patch) | |
tree | 2e87645c3175ed9a792d4957b65ee376934eb430 | |
parent | 76fc350395290f51ef6c0e6bbf5c97080eb5b644 (diff) | |
parent | c4e44426c9ee02275cbd075085526e8e334b16c5 (diff) |
Merge branch 'master' of ssh://git-annex.branchable.com
10 files changed, 192 insertions, 0 deletions
diff --git a/doc/bugs/adding_remote_server_using_ssh_on_a_4.1_device/comment_1_ab8a227df71c70c05074b50dcb798acd._comment b/doc/bugs/adding_remote_server_using_ssh_on_a_4.1_device/comment_1_ab8a227df71c70c05074b50dcb798acd._comment new file mode 100644 index 000000000..9edb91209 --- /dev/null +++ b/doc/bugs/adding_remote_server_using_ssh_on_a_4.1_device/comment_1_ab8a227df71c70c05074b50dcb798acd._comment @@ -0,0 +1,15 @@ +[[!comment format=mdwn + username="https://www.google.com/accounts/o8/id?id=AItOawnbPKLjBONawBd74MKJZo05juCqdsP1jAU" + nickname="Ramon" + subject="comment 1" + date="2015-02-27T01:43:42Z" + content=""" +I had the same problem, also with Android 4.1.1 (an Asus TF201). This is what I did: + +- Download version for Android 4.3 and 4.4 (autobuilds) and install that one over the previous. +- Add the remote, which worked just fine. +- Reinstall the version for Android 4.1 (autobuilds) + +And things seem to be working (things are syncing to the Android) + +"""]] diff --git a/doc/bugs/git.kitenet.net__47__downloads_has_wrong_git-annex_branch.mdwn b/doc/bugs/git.kitenet.net__47__downloads_has_wrong_git-annex_branch.mdwn new file mode 100644 index 000000000..c48c5201e --- /dev/null +++ b/doc/bugs/git.kitenet.net__47__downloads_has_wrong_git-annex_branch.mdwn @@ -0,0 +1,43 @@ +### Please describe the problem. + +This is coming from [[main repo not available on downloads.kitenet.net]], probably was the original issue, but since I already create a duplicate there, I won't assume anything (sorry about that!). + +### What steps will reproduce the problem? + +I am not sure. I know that I had an external drive with a clone of *some* repo related to `downloads.kitenet.net`. Doing a `git annex get` wouldn't work: + +<pre> +anarcat@marcos:current$ cd /media/anarcat/VHS/downloads.kitenet.net/git-annex/linux/current/ +anarcat@marcos:current$ git annex get git-annex-standalone-amd64.tar.gz +get git-annex-standalone-amd64.tar.gz (not available) + Try making some of these repositories available: + 840760dc-08f0-11e2-8c61-576b7e66acfd -- main repo + d7fa24ad-d104-4064-ad10-1078a4436e72 -- joey@elephant:~/lib/downloads +failed +git-annex: get: 1 failed +anarcat@marcos:current$ git remote -v +origin git://git.kitenet.net/downloads.git (fetch) +origin git://git.kitenet.net/downloads.git (push) +</pre> + +it turns out i somehow managed to checkout from `git://git.kitenet.net/downloads.git`. I don't know where I got this URL from, may it was back when I created that other bug report and the git URL wasn't explicitely mentionned on http://downloads.kitenet.net/ It turns out that there's a `git-annex` branch there that diverged from the "real" one, and is lacking tracking information. changing the remote here fixed the problem: + +<pre> +anarcat@marcos:current$ git remote set-url origin http://downloads.kitenet.net/.git/ +anarcat@marcos:current$ git remote update +Récupération de origin +Depuis http://downloads.kitenet.net/ + + e9febdc...1a80292 git-annex -> origin/git-annex (mise à jour forcée) +anarcat@marcos:current$ git annex get git-annex-standalone-amd64.tar.gz +get git-annex-standalone-amd64.tar.gz (from origin...) --2015-02-26 18:40:47-- http://downloads.kitenet.net/.git//annex/objects/Pm/8Z/SHA256E-s45589010--f82e2f600763b0f25a45a96ddf7ed68f26c67122e22cf8833d4c2473475bbce2.tar.gz/SHA256E-s45589010--f82e2f600763b0f25a45a96ddf7ed68f26c67122e22cf8833d4c2473475bbce2.tar.gz +Résolution de downloads.kitenet.net (downloads.kitenet.net)… 66.228.36.95, 2600:3c03::f03c:91ff:fe73:b0d2 +Connexion à downloads.kitenet.net (downloads.kitenet.net)|66.228.36.95|:80… connecté. +requête HTTP transmise, en attente de la réponse… 200 OK +[...] +</pre> + +### What version of git-annex are you using? On what operating system? + +5.20141125 on debian jessie. + +Sorry for the noise in that other bug report! I really thought it was the same issue .... --[[anarcat]] diff --git a/doc/forum/Android:_git_annex_sync_--content_not_run_in_all_repos__63__/comment_2_5b1565955ed88554e5ef4a00c0f4a754._comment b/doc/forum/Android:_git_annex_sync_--content_not_run_in_all_repos__63__/comment_2_5b1565955ed88554e5ef4a00c0f4a754._comment new file mode 100644 index 000000000..056c78547 --- /dev/null +++ b/doc/forum/Android:_git_annex_sync_--content_not_run_in_all_repos__63__/comment_2_5b1565955ed88554e5ef4a00c0f4a754._comment @@ -0,0 +1,8 @@ +[[!comment format=mdwn + username="https://www.google.com/accounts/o8/id?id=AItOawnbPKLjBONawBd74MKJZo05juCqdsP1jAU" + nickname="Ramon" + subject="comment 2" + date="2015-02-26T08:08:37Z" + content=""" +Thanks. I'll check as suggested if it happens again. Since switching between repos in android was complicated, I've started all over putting everything in a single repo (the default one). +"""]] diff --git a/doc/forum/Android:_unusably_slow_with_moderately_sized_repositories.mdwn b/doc/forum/Android:_unusably_slow_with_moderately_sized_repositories.mdwn new file mode 100644 index 000000000..ca2b3906f --- /dev/null +++ b/doc/forum/Android:_unusably_slow_with_moderately_sized_repositories.mdwn @@ -0,0 +1,67 @@ +I am trying to sync a repository between a computer and two Android devices. The data in the directory of files (i.e., not the .git directory) takes about 7.4 GB and there are about 10000 files and directories in total (as reported by find . | wc -l) of which 7000 are files (as reported by find - -type f | wc -l). + +I've used two different routes: 1. start the assistant in the Android in a directory with all the files: 2. start the assistant in the Android in an empty dir and sync with the git-annex repository in the computer. None are working. I am using, in the androids, v. 5.20150224-g9dca034 and in the computer the latest Debian version (5.20141125). + +Details follow: + + +Route 1. +========= + +In one Android, I had already transmitted all the files (nothing git-related; just the directory). So I placed that directory under /sdcard/annex, and started the assistant. I did not add any other repository (i.e., the computer was not added). After more than 12 hours, it is not done. The webapp in the Android shows "Adding blablabla". + +If I compare the sizes of directories between the computer and the android with du I get: + +Computer:~ du -H -d 1 +215624 ./.git +7710564 ./directory +7926192 . + + +Android (using adb shell):~ du -H -d 1 +7681744 ./directory +35388 ./.git +7717136 . + + +Over the last six hours it seems the size in the Android is getting closer and closer to that in the computer, but it seems is never quite getting there. And then, I think after this is completed (if ever), I'll need to sync with the computer. + + +Route 2. +======== + +In the other Android I started the assistant, and added the computer as the other repository. Let them sync. Initially I was getting the TransferScanner crashed problem in +[Android version does not sync](http://git-annex.branchable.com/forum/Android_version_does_not_sync/). So I issued the git-annex sync as suggested. Things started syncing. But again, after more than 12 hourse, it is not done and I have no idea how long I'd have to wait. In this case, git-annex seems to be placing stuff in ./git, as the things downloaded are actually links. A du shows this: + +392304 ./.git +82464 ./directory +474776 . + +------- + +I think I am experiencing two different problems: + +Problem 1. +========== + +git is just taking forever (route 1.) in the Android. There are not network issues possible there, since that Android is not downloading/syncing anything with anyone. Interestingly, in case this is of any use, from adb shell a top shows: + +User 27%, System 16%, IOW 1%, IRQ 0% +User 318 + Nice 28 + Sys 199 + Idle 674 + IOW 20 + IRQ 0 + SIRQ 0 = 1239 + + +and the git-annex process taking between 3% to 15% CPU. + + +Problem 2 +========== + +git + network issues in route 2. But I doubt this is mainly a network problem. As a comparison, using syncthing I was able to initially sync (i.e., download to the android device) the complete set of files in about four hours and then checking for changes in the Android takes syncthing about 1 to 3 minutes. I've also repeated this using a private seafile server (seafile server running in the computer); downloading all files takes about the same (about four hours), and then checking for changes against the seafile server takes between 45 minutes and an hour. For comparison, a top also shows that git-annex is taking between 2 and 15% CPU. + + +So I am wondering if I am making some silly mistake here, or maybe this is not the intended use case (because of number or size of files for an Android device). Of course, computer-to-computer things work just fine. It is the Android part which is not working. + + + + + diff --git a/doc/forum/Android:_unusably_slow_with_moderately_sized_repositories/comment_3_c5c655ee3ec2501c1c3c60fd83574914._comment b/doc/forum/Android:_unusably_slow_with_moderately_sized_repositories/comment_3_c5c655ee3ec2501c1c3c60fd83574914._comment new file mode 100644 index 000000000..42cab60b5 --- /dev/null +++ b/doc/forum/Android:_unusably_slow_with_moderately_sized_repositories/comment_3_c5c655ee3ec2501c1c3c60fd83574914._comment @@ -0,0 +1,8 @@ +[[!comment format=mdwn + username="https://www.google.com/accounts/o8/id?id=AItOawnbPKLjBONawBd74MKJZo05juCqdsP1jAU" + nickname="Ramon" + subject="comment 3" + date="2015-02-26T23:44:40Z" + content=""" +Please forgive the noise. I think I've been doing something very, very stupid. Sorry. I'd delete the whole thread if I could +"""]] diff --git a/doc/forum/Repository_backup/comment_2_e362221ef3d13bd417b2b6b0c51738a8._comment b/doc/forum/Repository_backup/comment_2_e362221ef3d13bd417b2b6b0c51738a8._comment new file mode 100644 index 000000000..e84be7f22 --- /dev/null +++ b/doc/forum/Repository_backup/comment_2_e362221ef3d13bd417b2b6b0c51738a8._comment @@ -0,0 +1,14 @@ +[[!comment format=mdwn + username="https://www.google.com/accounts/o8/id?id=AItOawnb2yfWzJ2lYQw1UTm6XVZ4y8qashNagZA" + nickname="Daniel" + subject="Re: comment1" + date="2015-02-26T12:48:56Z" + content=""" +Thanks for the tip! + +Yeah, I get that special remotes aren't meant for this, and backing up a bare (i.e. contentless) git repo somewhere else is pretty trivial. My only point was that it'd be nice to have a single unified offsite backup--if you backup the content to S3, say, and the git repo itself to Github, you now have two points of failure and, in some sense, twice the management hassle. (I now need both my Amazon and Github credentials to restore, for example.) + +This seems somewhat undesirable in a backup system. + +An easy way around this may be for me to just have a wrapper script that first backs up content to the special remote and then copies the .git repo (minus the objects) to the same place. So it's something I can easily enough work around. +"""]] diff --git a/doc/install/OSX/comment_10_e5172de344908f85ce6cf976e3c3806b._comment b/doc/install/OSX/comment_10_e5172de344908f85ce6cf976e3c3806b._comment new file mode 100644 index 000000000..99669d22e --- /dev/null +++ b/doc/install/OSX/comment_10_e5172de344908f85ce6cf976e3c3806b._comment @@ -0,0 +1,10 @@ +[[!comment format=mdwn + username="https://www.google.com/accounts/o8/id?id=AItOawk7iPiqWr3BVPLWEDvJhSSvcOqheLEbLNo" + nickname="Dirk" + subject="Is the Mavericks build still beeing updated?" + date="2015-02-26T12:22:51Z" + content=""" +I am still using Mac OS 10.9. Recently downloading a new git-annex version showed that the Mavericks build is still based on 5.20141104, while the Yosemite is 5.20150219. I am wondering if this is a misstake or if it is time for me to move on to Yosemite? ;-) + + +"""]] diff --git a/doc/tips/publishing_your_files_to_the_public/comment_4_37405f20da790141187e9f780c999448._comment b/doc/tips/publishing_your_files_to_the_public/comment_4_37405f20da790141187e9f780c999448._comment new file mode 100644 index 000000000..2855c3fdd --- /dev/null +++ b/doc/tips/publishing_your_files_to_the_public/comment_4_37405f20da790141187e9f780c999448._comment @@ -0,0 +1,11 @@ +[[!comment format=mdwn + username="BojanNikolic" + subject="comment 4" + date="2015-02-27T09:55:15Z" + content=""" +Many thanks. The command line I ended up using is: + + fname=\"2015/01/04/myfile.pdf\" ; git annex copy --to pubweb $fname; key=`git annex lookupkey \"$fname\"`; git annex examinekey $key --format 'https://www.myweb.com/d/${hashdirlower}${key}/${key}\n' + +to publish selected documents in my git-annex repository onto the web via a rsync special remote on a conventional http server. +"""]] diff --git a/doc/todo/server-level_daemon__63__/comment_2_4a5ccb0e772352adbf20940c010895a5._comment b/doc/todo/server-level_daemon__63__/comment_2_4a5ccb0e772352adbf20940c010895a5._comment new file mode 100644 index 000000000..cc6ab9848 --- /dev/null +++ b/doc/todo/server-level_daemon__63__/comment_2_4a5ccb0e772352adbf20940c010895a5._comment @@ -0,0 +1,9 @@ +[[!comment format=mdwn + username="https://id.koumbit.net/anarcat" + subject="comment 2" + date="2015-02-26T23:52:52Z" + content=""" +because this is a server without a generic \"user account\". there are multiple admins with each their own account, and i'd like to have git-annex run as its own account. and yes, probably under systemd, but under wheezy, under init.d. + +i would avoid @reboot cron jobs because systemd can't restart it if they fail. +"""]] diff --git a/doc/todo/server-level_daemon__63__/comment_3_60c5030a21c28b79f634f0655bc6c05d._comment b/doc/todo/server-level_daemon__63__/comment_3_60c5030a21c28b79f634f0655bc6c05d._comment new file mode 100644 index 000000000..974cc0a11 --- /dev/null +++ b/doc/todo/server-level_daemon__63__/comment_3_60c5030a21c28b79f634f0655bc6c05d._comment @@ -0,0 +1,7 @@ +[[!comment format=mdwn + username="https://id.koumbit.net/anarcat" + subject="comment 3" + date="2015-02-26T23:53:32Z" + content=""" +why would that be a bad idea anyways? +"""]] |