afpfs-ng-devel Mailing List for afpfs-ng (Page 6)
Status: Alpha
Brought to you by:
alexthepuffin
You can subscribe to this list here.
2006 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(7) |
Dec
(10) |
---|---|---|---|---|---|---|---|---|---|---|---|---|
2007 |
Jan
(8) |
Feb
(9) |
Mar
(16) |
Apr
(1) |
May
(19) |
Jun
(5) |
Jul
(4) |
Aug
(6) |
Sep
(7) |
Oct
(14) |
Nov
|
Dec
|
2008 |
Jan
|
Feb
(4) |
Mar
(14) |
Apr
(6) |
May
(2) |
Jun
(14) |
Jul
(11) |
Aug
(2) |
Sep
|
Oct
|
Nov
|
Dec
|
2009 |
Jan
(3) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
2011 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
(2) |
Nov
|
Dec
|
2012 |
Jan
(3) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
From: Alex d. <ale...@gm...> - 2007-03-29 14:21:41
|
Here's what I understand: - you have an LDAP server running on Mac OS X - you have Linux clients that use that LDAP server - you want your users to login once to the Linux machines using their LDAP authentication, and be able to see their AFP shares - Alex On 3/29/07, dho...@ha... <dho...@ha...> wrote: > Quoting Alex deVries <ale...@gm...>: > > > On 3/28/07, dho...@ha... > > <dho...@ha...> wrote: > >> > >> >> Do I need to define a mount point in fstab? > > > > Actually, you can't. Entries in /etc/fstab are mounted as root, and > > right now an afpfsd instance is on a per-user basis. This means that > > if you have 10 different users logged in, there need to be 10 > > different copies of afpfsd. afp_client will start one up if it hasn't > > already started. So adding it to rc.local won't really help. > > > > I'll change that at some point. Mapping local to remote users is a > > bit complicated, and there are some bugs in 0.4 in that area. > > > > Here's a question for you: all afp mounts require some form of > > authentication. What local and remote user accounts would you use for > > that mount? Is it shared between users? > > > > Glad you like it. Tell your friends and it'll get better. > > > > - Alex > > The mount is not shared between users. Authentication is currently > through Apples Active Directory (Openldap) user accounts created > through workgroup manager. I need to make this as simple as possible > for the user. Right now the workstation has a generic account and a > root account. Should I create an account with the same info as user > account on the apple server? > > -- > Dwayne Hottinger > Network Administrator > Harrisonburg City Public Schools > > |
From: <dho...@ha...> - 2007-03-29 12:28:50
|
Quoting Alex deVries <ale...@gm...>: > On 3/28/07, dho...@ha... > <dho...@ha...> wrote: >> >> >> Do I need to define a mount point in fstab? > > Actually, you can't. Entries in /etc/fstab are mounted as root, and > right now an afpfsd instance is on a per-user basis. This means that > if you have 10 different users logged in, there need to be 10 > different copies of afpfsd. afp_client will start one up if it hasn't > already started. So adding it to rc.local won't really help. > > I'll change that at some point. Mapping local to remote users is a > bit complicated, and there are some bugs in 0.4 in that area. > > Here's a question for you: all afp mounts require some form of > authentication. What local and remote user accounts would you use for > that mount? Is it shared between users? > > Glad you like it. Tell your friends and it'll get better. > > - Alex The mount is not shared between users. Authentication is currently =20 through Apples Active Directory (Openldap) user accounts created =20 through workgroup manager. I need to make this as simple as possible =20 for the user. Right now the workstation has a generic account and a =20 root account. Should I create an account with the same info as user =20 account on the apple server? --=20 Dwayne Hottinger Network Administrator Harrisonburg City Public Schools |
From: Alex d. <ale...@gm...> - 2007-03-29 12:13:50
|
On 3/28/07, dho...@ha... <dho...@ha...> wrote: > > >> Do I need to define a mount point in fstab? Actually, you can't. Entries in /etc/fstab are mounted as root, and right now an afpfsd instance is on a per-user basis. This means that if you have 10 different users logged in, there need to be 10 different copies of afpfsd. afp_client will start one up if it hasn't already started. So adding it to rc.local won't really help. I'll change that at some point. Mapping local to remote users is a bit complicated, and there are some bugs in 0.4 in that area. Here's a question for you: all afp mounts require some form of authentication. What local and remote user accounts would you use for that mount? Is it shared between users? Glad you like it. Tell your friends and it'll get better. - Alex |
From: <dho...@ha...> - 2007-03-29 00:47:36
|
Quoting Alex deVries <ale...@gm...>: > Dwayne, > > Give this a try: > > Start by killing off afpfsd (kill -9 afpfsd). > > In one window, run 'afpfsd -d', and leave it there. > > In another window, run 'afp_client mount ...' > > If you see: > Permission denied, maybe a problem with the fuse device or mountpoint? > It means that you have incorrect permissions on /usr/bin/fusermount > (chmod 4755 /usr/bin/fusermount will fix this) > > If you see: > fusermount: failed to open /dev/fuse: Permission denied > > It means that you don't have permissions on /dev/fuse. chmod > /dev/fuse will fix that too. > > And if that doesn't work, let me know. > Almost got it. I created /media/pear. Did afp_client mount =20 pear:username-u username-p password /media/pear said mounted it =20 successfully. However in the other tab I have the following errors: Starting up AFPFS version 0.4 Got connection 4 mounting dhottinger on /pear Got a signal 12 Got a server signal, 0 Got a signal 12 Got a server signal, 0 Got a signal 12 Got a server signal, 0 Got a signal 12 Got a server signal, 0 Removing connection for server pear Got a signal 12 Got a server signal, 0 Starting mount. Creating new connection to server Got a signal 12 Got a server signal, 0 Actually mounting. fuse: bad mount point `/pear': No such file or directory Permission denied, maybe a problem with the fuse device or mountpoint? Removing connection for server pear Got a signal 12 Got a server signal, 0 Got a signal 12 Got a server signal, 0 reading: Interrupted system call Got a signal 12 Got a server signal, 0 Got connection 4 mounting dhottinger on /pear Got a signal 12 Got a server signal, 0 Got a signal 12 Got a server signal, 0 Got a signal 12 Got a server signal, 0 Removing connection for server pear Got a signal 12 Got a server signal, 0 Starting mount. Creating new connection to server Got a signal 12 Got a server signal, 0 Actually mounting. fuse: bad mount point `/pear': No such file or directory Permission denied, maybe a problem with the fuse device or mountpoint? Removing connection for server pear Got a signal 12 Got a server signal, 0 Got a signal 12 Got a server signal, 0 Got a signal 12 Got a server signal, 0 Got connection 4 mounting dhottinger on /pear Got a signal 12 Got a server signal, 0 Got a signal 12 Got a server signal, 0 Got a signal 12 Got a server signal, 0 Removing connection for server pear Got a signal 12 Got a server signal, 0 Starting mount. Creating new connection to server Got a signal 12 Got a server signal, 0 Actually mounting. fuse: bad mount point `/pear': No such file or directory Permission denied, maybe a problem with the fuse device or mountpoint? Removing connection for server pear Got a signal 12 Got a server signal, 0 Got a signal 12 Got a server signal, 0 reading: Interrupted system call Got a signal 12 Got a server signal, 0 Got connection 4 mounting dhottinger on /media/pear Got a signal 12 Got a server signal, 0 Got a signal 12 Got a server signal, 0 Got a signal 12 Got a server signal, 0 Removing connection for server pear Got a signal 12 Got a server signal, 0 Starting mount. Creating new connection to server Got a signal 12 Got a server signal, 0 Actually mounting. unique: 1, opcode: INIT (26), nodeid: 0, insize: 56 INIT: 7.7 flags=3D0x00000003 max_readahead=3D0x00020000 Mounting succeeded. reading: Interrupted system call Got a signal 12 Got a server signal, 0 INIT: 7.8 flags=3D0x00000001 max_readahead=3D0x00020000 max_write=3D0x00020000 unique: 1, error: 0 (Success), outsize: 40 unique: 2, opcode: GETATTR (3), nodeid: 1, insize: 40 unique: 2, error: -5 (Input/output error), outsize: 16 unique: 3, opcode: GETATTR (3), nodeid: 1, insize: 40 unique: 3, error: -5 (Input/output error), outsize: 16 unique: 4, opcode: GETATTR (3), nodeid: 1, insize: 40 unique: 4, error: -5 (Input/output error), outsize: 16 unique: 5, opcode: GETATTR (3), nodeid: 1, insize: 40 unique: 5, error: -5 (Input/output error), outsize: 16 unique: 6, opcode: GETATTR (3), nodeid: 1, insize: 40 unique: 6, error: -5 (Input/output error), outsize: 16 Got connection 4 mounting dhottinger on /media/pear Actually mounting. Volume dhottinger is already mounted Got a signal 12 Got a server signal, 0 reading: Interrupted system call Got a signal 12 Got a server signal, 0 unique: 7, opcode: GETATTR (3), nodeid: 1, insize: 40 unique: 7, error: -5 (Input/output error), outsize: 16 unique: 8, opcode: LOOKUP (1), nodeid: 1, insize: 51 LOOKUP /dhottinger unique: 8, error: -2 (No such file or directory), outsize: 16 unique: 9, opcode: LOOKUP (1), nodeid: 1, insize: 51 LOOKUP /dhottinger unique: 9, error: -2 (No such file or directory), outsize: 16 Doing a ls /media/pear returns an i/o error. So I did an exit, and restarted the process. Created a directory =20 called pear remounted everything. Looks good. Did ls /pear and I get =20 [root@linux-compaq pear]# ls /pear ls: /pear: Input/output error In second tab I have: Starting mount. Creating new connection to server Got a signal 12 Got a server signal, 0 Actually mounting. unique: 1, opcode: INIT (26), nodeid: 0, insize: 56 INIT: 7.7 flags=3D0x00000003 max_readahead=3D0x00020000 Mounting succeeded. reading: Interrupted system call Got a signal 12 Got a server signal, 0 INIT: 7.8 flags=3D0x00000001 max_readahead=3D0x00020000 max_write=3D0x00020000 unique: 1, error: 0 (Success), outsize: 40 unique: 2, opcode: GETATTR (3), nodeid: 1, insize: 40 unique: 2, error: -5 (Input/output error), outsize: 16 unique: 3, opcode: GETATTR (3), nodeid: 1, insize: 40 unique: 3, error: -5 (Input/output error), outsize: 16 unique: 4, opcode: GETATTR (3), nodeid: 1, insize: 40 unique: 4, error: -5 (Input/output error), outsize: 16 unique: 5, opcode: GETATTR (3), nodeid: 1, insize: 40 unique: 5, error: -5 (Input/output error), outsize: 16 unique: 6, opcode: GETATTR (3), nodeid: 1, insize: 40 unique: 6, error: -5 (Input/output error), outsize: 16 unique: 7, opcode: GETATTR (3), nodeid: 1, insize: 40 unique: 7, error: -5 (Input/output error), outsize: 16 Maybe a flag I need to pass? I have netatalk running on this =20 workstation. Incidently Right now I am running all these commands as =20 root. thanks, ddh --=20 Dwayne Hottinger Network Administrator Harrisonburg City Public Schools |
From: <dho...@ha...> - 2007-03-29 00:24:20
|
Quoting Alex deVries <ale...@gm...>: > Dwayne, > > Give this a try: > > Start by killing off afpfsd (kill -9 afpfsd). > > In one window, run 'afpfsd -d', and leave it there. > > In another window, run 'afp_client mount ...' > > If you see: > Permission denied, maybe a problem with the fuse device or mountpoint? > It means that you have incorrect permissions on /usr/bin/fusermount > (chmod 4755 /usr/bin/fusermount will fix this) > > If you see: > fusermount: failed to open /dev/fuse: Permission denied > > It means that you don't have permissions on /dev/fuse. chmod > /dev/fuse will fix that too. > > And if that doesn't work, let me know. > > > - Alex > > > > On 3/28/07, dho...@ha... > <dho...@ha...> wrote: >> I installed afpfs-ng-0.4 On a fresh Fedora Core 6 box. When I try to >> mount my user volume on an OS X 10.4 server I get an error message: >> Starting mount. >> Creating new connection to server >> Actually mounting. >> Permission denied, maybe a problem with the fuse device or mountpoint? >> reading: Interrupted system call >> >> Do I need to define a mount point in fstab? Thanks, I'll give it a try and let you know. Cool software. Its going to =20 save me a lot of headache if I can get it to work. Im thinking I =20 should be able to script it out in rc.local. --=20 Dwayne Hottinger Network Administrator Harrisonburg City Public Schools |
From: Alex d. <ale...@gm...> - 2007-03-28 22:23:49
|
Dwayne, Give this a try: Start by killing off afpfsd (kill -9 afpfsd). In one window, run 'afpfsd -d', and leave it there. In another window, run 'afp_client mount ...' If you see: Permission denied, maybe a problem with the fuse device or mountpoint? It means that you have incorrect permissions on /usr/bin/fusermount (chmod 4755 /usr/bin/fusermount will fix this) If you see: fusermount: failed to open /dev/fuse: Permission denied It means that you don't have permissions on /dev/fuse. chmod /dev/fuse will fix that too. And if that doesn't work, let me know. - Alex On 3/28/07, dho...@ha... <dho...@ha...> wrote: > I installed afpfs-ng-0.4 On a fresh Fedora Core 6 box. When I try to > mount my user volume on an OS X 10.4 server I get an error message: > Starting mount. > Creating new connection to server > Actually mounting. > Permission denied, maybe a problem with the fuse device or mountpoint? > reading: Interrupted system call > > Do I need to define a mount point in fstab? > > ddh > > -- > Dwayne Hottinger > Network Administrator > Harrisonburg City Public Schools > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share your > opinions on IT & business topics through brief surveys-and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > Afpfs-ng-devel mailing list > Afp...@li... > https://lists.sourceforge.net/lists/listinfo/afpfs-ng-devel > |
From: <dho...@ha...> - 2007-03-28 22:15:38
|
I installed afpfs-ng-0.4 On a fresh Fedora Core 6 box. When I try to mount my user volume on an OS X 10.4 server I get an error message: Starting mount. Creating new connection to server Actually mounting. Permission denied, maybe a problem with the fuse device or mountpoint? reading: Interrupted system call Do I need to define a mount point in fstab? ddh -- Dwayne Hottinger Network Administrator Harrisonburg City Public Schools |
From: Michael G. <mi...@li...> - 2007-02-13 19:38:35
|
Alex deVries schrieb: > Michael, > > It would help me out quite a bit if you could test running afpfs-ng on > Debian, and make Debian packages for it. > do you need some 8.6 and/or 9.2 cds for imac? > - Alex > > On 2/13/07, Michael Grundmann <mi...@li...> wrote: >> Hello, >> >> let me know how i can help!! >> >> I'm a Macuser since 1994 - and i can have some netatalk, helios and os >> x-server here for testing. >> Ok - my english is _very_ bad, but i will do my very best. >> My servers are running with debian - so i can help with dep-files. >> >> For special-test-need send some postings. >> >> Michael >> >> ------------------------------------------------------------------------- >> >> Using Tomcat but need to do more? Need to support web services, >> security? >> Get stuff done quickly with pre-integrated technology to make your >> job easier. >> Download IBM WebSphere Application Server v.1.0.1 based on Apache >> Geronimo >> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 >> _______________________________________________ >> Afpfs-ng-devel mailing list >> Afp...@li... >> https://lists.sourceforge.net/lists/listinfo/afpfs-ng-devel >> > |
From: Alex d. <ale...@gm...> - 2007-02-13 18:26:53
|
Michael, It would help me out quite a bit if you could test running afpfs-ng on Debian, and make Debian packages for it. - Alex On 2/13/07, Michael Grundmann <mi...@li...> wrote: > Hello, > > let me know how i can help!! > > I'm a Macuser since 1994 - and i can have some netatalk, helios and os > x-server here for testing. > Ok - my english is _very_ bad, but i will do my very best. > My servers are running with debian - so i can help with dep-files. > > For special-test-need send some postings. > > Michael > > ------------------------------------------------------------------------- > Using Tomcat but need to do more? Need to support web services, security? > Get stuff done quickly with pre-integrated technology to make your job easier. > Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 > _______________________________________________ > Afpfs-ng-devel mailing list > Afp...@li... > https://lists.sourceforge.net/lists/listinfo/afpfs-ng-devel > |
From: Michael G. <mi...@li...> - 2007-02-13 17:43:00
|
Hello, let me know how i can help!! I'm a Macuser since 1994 - and i can have some netatalk, helios and os x-server here for testing. Ok - my english is _very_ bad, but i will do my very best. My servers are running with debian - so i can help with dep-files. For special-test-need send some postings. Michael |
From: Alex d. <ale...@gm...> - 2007-02-12 21:42:06
|
Okay, give this a try. Clearly I'm not detecting the FUSE version correctly. BTW, if anyone out there wants to generate any Debian, Ubuntu, Gentoo, Slackware, whatever patches, I would really appreciate it. - Alex On 2/12/07, Martin Forsgren <mk...@st...> wrote: > Hi > I cannot compile this version. > What can I try? > > Output from make is attached. > > /Martin > > --------------------------------------- > Debian Testing > libfuse2 2.5.3-4.1 > libfuse-dev 2.5.3-4.1 > libgcrypt11 1.2.3-2 > libgcrypt11-dev 1.2.3-2 > gcc 4.1.1-15 > ------------------------------------------------------------------------- > Using Tomcat but need to do more? Need to support web services, security? > Get stuff done quickly with pre-integrated technology to make your job easier. > Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 > _______________________________________________ > Afpfs-ng-devel mailing list > Afp...@li... > https://lists.sourceforge.net/lists/listinfo/afpfs-ng-devel > > > |
From: Martin F. <mk...@st...> - 2007-02-12 13:06:39
|
Hi I cannot compile this version. What can I try? Output from make is attached. /Martin --------------------------------------- Debian Testing libfuse2 2.5.3-4.1 libfuse-dev 2.5.3-4.1 libgcrypt11 1.2.3-2 libgcrypt11-dev 1.2.3-2 gcc 4.1.1-15 |
From: Alex d. <ale...@gm...> - 2007-02-12 02:14:45
|
We're happy to release version 0.4 of afpfs-ng, the Apple Filing Protocol client for Linux with FUSE. New features include: - stability: afpfs-ng is reliable to the point of being usable for all but the most strenuous IO loads - encrypted authentication mechanisms: now also supporting Randnum, 2-way randnum, DHCAST128 and DHX2 - performance: DID caching, enhanced DSI packet processing and other improvements make afpfs-ng about half as fast as the Mac OS X client - ease of use: unmounting, SIGINT, easier logging, easier startup, better docs, session suspension For more information and to download, see http://afpfs-ng.sf.net/. afpfs-ng will let you mount remove AFP volumes from Mac OS X or netatalk on a Linux system (or any system that has FUSE). - Alex |
From: Alex d. <ale...@gm...> - 2007-02-05 03:51:46
|
To make merging a little easier, afpfs-ng is now in CVS. You should be able to get it with: cvs -d :pserver:ano...@af...:/cvsroot/afpfs-ng co afpfs-ng What's in there now is well on its way to being 0.4. Key things that have change since 0.3.2: - Integrated various advanced UAMs (thanks, Derrik Pates) - huge memory leaks fixed - overall stability; you can now untar and build a Linux kernel - could be 64-bit safe (untested) - unmounting works properly - suspend/resume connections - various fixes to chmod, readdir, write - start of uid/gid mapping (where what's on the server's not what you have locally) - DSI and afpRead bugs fixed - implemnted a DID cache to improve performance (currently disabled) - fixes to meta data (icons, comments, forks) - lots, lots more The largest obstacle before 0.4 are: - it leaks gobs and gobs of memory on write - updated documentation - packages for distributions (FC6, Ubuntu, Debian) Any help is appreciated... - Alex |
From: Alex d. <ale...@gm...> - 2007-02-01 03:03:37
|
On 1/31/07, Derrik Pates <de...@de...> wrote: ... > After doing this, I could successfully authenticate against a MacOS X 10.4 > system at the office using the DHCAST128 UAM. I've also transcribed my > Randnum and 2-way Randnum UAMs, but I don't currently have a means to > test them. Sounds great! Could you mail me a patch? > Also, it seems that the function that should be returning the dirent list > for a directory is only getting a small fraction of the entries; I suspect > that FPEnumerate{,Ext,Ext2} (whichever you're using) is only getting called > one time; you need to call the enumerator in a loop, until you don't get > any more entries, otherwise with anything other than a very small > directory, entries will be missed. That's interesting... I haven't seen that behaviour since 0.3.2. afp_readdir() in fuse_int.c has such a loop. I just tested it with 1000 dir entries and I don't see that behaviour right now. Can you describe your setup? - Alex |
From: Derrik P. <de...@de...> - 2007-02-01 02:44:04
|
Alex: After some diagnosing, I discovered that the password length was hardcoded in your 'afp_server_mount_request' structure at 9 characters (obviously 8 characters plus a trailing \0). Unfortunately that means that any time you have a longer password, you can never authenticate, as it always gets truncated. (The DHCAST128 UAM supports passwords up to 64 characters; DHX2 on the other hand supports passwords up to 255 characters.) After lengthening the password field in the structure definition, and altering the snprintf() invocation in client.c to allow a password of up to 255 characters, nothing worked - it appears the structure was long enough that it broke the whole system. (The check in process_mount(), comparing len to sizeof(struct afp_server_mount_request), fails.) After changing the username and password field lengths to 127 characters each, it started working again. After doing this, I could successfully authenticate against a MacOS X 10.4 system at the office using the DHCAST128 UAM. I've also transcribed my Randnum and 2-way Randnum UAMs, but I don't currently have a means to test them. Also, it seems that the function that should be returning the dirent list for a directory is only getting a small fraction of the entries; I suspect that FPEnumerate{,Ext,Ext2} (whichever you're using) is only getting called one time; you need to call the enumerator in a loop, until you don't get any more entries, otherwise with anything other than a very small directory, entries will be missed. -- Derrik Pates de...@de... |
From: Alex d. <ale...@gm...> - 2007-01-31 15:22:18
|
On 1/31/07, Derrik Pates <de...@de...> wrote: > I'll have to see what I can do about debugging that as well. My > implementation has been tested primarily against OS X 10.3 and 10.4, > but minor tweaks allowed me to have it work against Netatalk. I've > not attempted building a kernel (though that would be an interesting > test), but I've copied files to and from AFP mounts, including playback > of video and audio at full speed over 100 Mbps Ethernet. I've also > implemented extensive caching of dirent lists and stat data, as well as > symbolic link targets. For being pure Perl, it works quite well. I found that doing simple streaming of reads and writes worked well enough. Simultaneous reads and writes brought up a lot of problems with my DSI packet handling, particularly when running multithreaded. I'm not sure what kind of license you have on your code, but I would like to see your perl implementation... - A |
From: Derrik P. <de...@de...> - 2007-01-31 15:07:52
|
On Wed, Jan 31, 2007 at 09:54:00AM -0500, Alex deVries wrote: > Sounds great! I'd love to see your code. Doing this alone is getting > a bit tiring. > > I have actually rewritten the UAMs interface with the expectation of > writing new UAM support, but I haven't written any new ones. As you > mentioned, clearly the old interface was lacking. I actually started last night with retooling the UAM interface and writing an implementation of the "Randnum Exchange" UAM translated from my Perl version. > So as to facilitate developement, I've just pumped out 0.3.3 which you > can download. You'll be able to see my UAMs interface. Now, if this > isn't a sufficiently flexible interface, I have no problems whatsoever > in changing it. Just send me a patch. I'll pull it down tonight and see how the new interface looks. > But, I have succesfully been able to copy a kernel tarball, untar it > and compile it with a loopback connection to netatalk. I can get part > of the way with a mount to an OSX machine. I'll have to see what I can do about debugging that as well. My implementation has been tested primarily against OS X 10.3 and 10.4, but minor tweaks allowed me to have it work against Netatalk. I've not attempted building a kernel (though that would be an interesting test), but I've copied files to and from AFP mounts, including playback of video and audio at full speed over 100 Mbps Ethernet. I've also implemented extensive caching of dirent lists and stat data, as well as symbolic link targets. For being pure Perl, it works quite well. -- Derrik Pates de...@de... |
From: Alex d. <ale...@gm...> - 2007-01-31 14:54:07
|
Derrik, Sounds great! I'd love to see your code. Doing this alone is getting a bit tiring. I have actually rewritten the UAMs interface with the expectation of writing new UAM support, but I haven't written any new ones. As you mentioned, clearly the old interface was lacking. So as to facilitate developement, I've just pumped out 0.3.3 which you can download. You'll be able to see my UAMs interface. Now, if this isn't a sufficiently flexible interface, I have no problems whatsoever in changing it. Just send me a patch. And a note about 0.3.3... this is a release which is much better than 0.3.2, but it has some known problems, such as: - leaking memory on writes (I know the problem now) - on concurrent reads and writes, there are some problems where the accelerated dsi packet processing gets out of sync. I've solved about 10 of these problems in the last week. - I know it has problems with fuse 2.6.1 (with threading, a problem in fuse I think), and is untested against fuse 2.5.3. Works with 2.6.0. But, I have succesfully been able to copy a kernel tarball, untar it and compile it with a loopback connection to netatalk. I can get part of the way with a mount to an OSX machine. I'd really prefer that people hang on until version 0.4 before using afpfs-ng, although I'm at the point where quality is converging (ie. I'm fixing bugs at a faster rate than I'm finding them). - Alex On 1/30/07, Derrik Pates <de...@de...> wrote: > All, > > I've actually progressively been writing my own AFP library, as well as a > FUSE-based tool to mount an AFP (over TCP) filesystem on Linux - however, > I wrote mine completely in Perl. I'd thought to start adapting my UAM > code to afpfs-ng - I have implemented UAMs for both 1- and 2-way random > number auth, as well as DHCAST128 and DHX2 UAMs. > > Unfortunately the existing interface for UAMs is not sufficient for > anything other than plain-text authentication (which is, by the way, > limited to 8-character passwords). Having a function that just generates > an authinfo block isn't enough for UAMs like DHX2, where 'FPLogin' is > called, followed by two different invocations of 'FPLoginCont'. I can > provide my implementations as examples, but the preferred way would be to > have one function that accepts the username and password (and AFP version, > if needed - it might already be indicated in the AFP session object). > > I've begun reading the libgcrypt API documentation, which has the necessary > interfaces for DES and CAST5 encryption, CBC, and MD5 hashing, but > without retooling the UAM interface, it won't be possible to practically > implement the more advanced UAMs. > > -- > Derrik Pates > de...@de... > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share your > opinions on IT & business topics through brief surveys - and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > Afpfs-ng-devel mailing list > Afp...@li... > https://lists.sourceforge.net/lists/listinfo/afpfs-ng-devel > |
From: Derrik P. <de...@de...> - 2007-01-31 04:47:00
|
All, I've actually progressively been writing my own AFP library, as well as a FUSE-based tool to mount an AFP (over TCP) filesystem on Linux - however, I wrote mine completely in Perl. I'd thought to start adapting my UAM code to afpfs-ng - I have implemented UAMs for both 1- and 2-way random number auth, as well as DHCAST128 and DHX2 UAMs. Unfortunately the existing interface for UAMs is not sufficient for anything other than plain-text authentication (which is, by the way, limited to 8-character passwords). Having a function that just generates an authinfo block isn't enough for UAMs like DHX2, where 'FPLogin' is called, followed by two different invocations of 'FPLoginCont'. I can provide my implementations as examples, but the preferred way would be to have one function that accepts the username and password (and AFP version, if needed - it might already be indicated in the AFP session object). I've begun reading the libgcrypt API documentation, which has the necessary interfaces for DES and CAST5 encryption, CBC, and MD5 hashing, but without retooling the UAM interface, it won't be possible to practically implement the more advanced UAMs. -- Derrik Pates de...@de... |
From: Martin F. <mk...@st...> - 2007-01-04 17:15:18
|
Looking forward for next release! Just for curiosity some questions about this: Is the protocols used good? Is it open (can you find the specs somewhere)? How is the Microsoft alternative (broadcast services and alow to connec= t = to them)? Is there any other alternatives working? Keep it up /Martin = Den 2007-01-02 19:45:50 skrev Alex deVries <ale...@gm...>: > afpfs-ng 0.3.2 is not quite functional with OSX, and this is something= > I know about and am working on. > > Most of my work's been against netatalk, which behaves quite > differently from OS X. An example of such a difference is the error > codes that are returned on some operations. Existing afpfs-ng > behaviour will kill the connection. There were also many problems > with large directory listings, file permissions, etc, etc. 10.3 is > also quite different from 10.4. > > Another feature that's better is connection error handling. The use > case that Martin's brought up is being able to connect to a share, > close your laptop and walk away, open up the laptop on the same > network again and see the same share as if you'd never left. I have > some of this working. > > I don't want people to get the impression that I've walked away from > afpfs-ng or anything, I'm just knee-deep in the code. > > And if anyone out there has a copy of Mac OS 8.6 or 9.2 for pre-iMac > PPC macs, I'd really appreciate a copy so I can do some AFP 2.2 > development. > > - Alex > > ----------------------------------------------------------------------= --- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to shar= e = > your > opinions on IT & business topics through brief surveys - and earn cash= > http://www.techsay.com/default.php?page=3Djoin.php&p=3Dsourceforge&CID= =3DDEVDEV > _______________________________________________ > Afpfs-ng-devel mailing list > Afp...@li... > https://lists.sourceforge.net/lists/listinfo/afpfs-ng-devel |
From: Houritsu C. <hou...@ho...> - 2007-01-04 16:25:58
|
Hi, Alex. > Did you have that problem with 0.3 also? No, I found that with only ver.0.3.2. With 0.3, I can read some directory, but I can't read other directory. However, I don't think that is prior problem. At first, you do something which you want to do.> BTW, do you want me to distribute your .spec file with the afpfs-ng 0.4? If you want to include .spec file in afpfs-ng archive, please do it. But that .spec file depends on my patch which causes the program afp_client, afpfsd to be installed in '/usr/sbin'.Because other network daemons in Fedora are installed in '/usr/sbin'. Then I think you should remove that patch line in spec file when you include that in archive.-- 法律厨 hou...@ho...> Date: Wed, 3 Jan 2007 20:08:03 -0500> From: ale...@gm...> To: hou...@ho...> Subject: Re: [Afpfs-ng-devel] rpm of afpfs-ng> > Houritsu-san,> > Did you have that problem with 0.3 also?> > There are pretty large problems with the way that 0.3 reads files; 0.4> will be out in a few weeks and it'll address this problem.> > BTW, do you want me to distribute your .spec file with the afpfs-ng 0.4?> > - Alex> > On 1/3/07, Houritsu Chu <hou...@ho...> wrote:> >> >> > Hi, developers.> >> > I remake afpfs-ng RPM of afpfs-ng ver.0.3.2, and> > try it.But result is not so good.afpfs-ng success to> > mount netatalk volume, but cannot read subdirectories.> >> > My RPM is on http://ohkawa.biz/Linux/.> >> > Test Environment:> > Afpfs-ng:> > CPU of PC: AMD AthlonXP 3400+:> > Memory:128MB> > OS:Fedora Core 6(x86_64)> >> > Netatalk:> > CPU of PC:Intel Celeron 2.1GHz> > Memory: 192MB> > OS:Fedora Core 6(i386)> >> > --> >> > houritsuchu> >> > hou...@ho...> >> > ________________________________> > 最新の Wi ndows Live メッセンジャー。 今すぐトライ!> > -------------------------------------------------------------------------> > Take Surveys. Earn Cash. Influence the Future of IT> > Join SourceForge.net's Techsay panel and you'll get the chance to share your> > opinions on IT & business topics through brief surveys - and earn cash> > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV> >> > _______________________________________________> > Afpfs-ng-devel mailing list> > Afp...@li...> > https://lists.sourceforge.net/lists/listinfo/afpfs-ng-devel> >> >> > _________________________________________________________________ 「セキュリティ+PC メンテナンス」を提供する All in One PC ケアの Windows Live OneCare ベータ版 (無料) を今すぐトライ! http://www.ideas.live.com/programpage.aspx?versionid=b2456790-90e6-4d28-9219-5d7207d94d45 |
From: Houritsu C. <hou...@ho...> - 2007-01-03 22:54:13
|
Hi, developers.I remake afpfs-ng RPM of afpfs-ng ver.0.3.2, andtry it.But result is not so good.afpfs-ng success to mount netatalk volume, but cannot read subdirectories. My RPM is on http://ohkawa.biz/Linux/.Test Environment:Afpfs-ng:CPU of PC: AMD AthlonXP 3400+:Memory:128MBOS:Fedora Core 6(x86_64)Netatalk:CPU of PC:Intel Celeron 2.1GHzMemory: 192MBOS:Fedora Core 6(i386)-- houritsuchu hou...@ho... _________________________________________________________________ 最新の Windows Live メッセンジャーを今すぐトライ! http://promotion.live.jp/messenger/ |
From: Alex d. <ale...@gm...> - 2007-01-02 18:45:56
|
afpfs-ng 0.3.2 is not quite functional with OSX, and this is something I know about and am working on. Most of my work's been against netatalk, which behaves quite differently from OS X. An example of such a difference is the error codes that are returned on some operations. Existing afpfs-ng behaviour will kill the connection. There were also many problems with large directory listings, file permissions, etc, etc. 10.3 is also quite different from 10.4. Another feature that's better is connection error handling. The use case that Martin's brought up is being able to connect to a share, close your laptop and walk away, open up the laptop on the same network again and see the same share as if you'd never left. I have some of this working. I don't want people to get the impression that I've walked away from afpfs-ng or anything, I'm just knee-deep in the code. And if anyone out there has a copy of Mac OS 8.6 or 9.2 for pre-iMac PPC macs, I'd really appreciate a copy so I can do some AFP 2.2 development. - Alex |
From: Alex d. <ale...@gm...> - 2006-12-16 23:45:00
|
I'm looking into trying to have proper UTF8 support. I've never done this before. Does anyone have any suggestions on what to use for characater translation? In looking through netatalk, they have their own copy of iconv with some Mac encodings; there's the following note: " From samba 3.0 beta and GNU libiconv-1.8 It's bad but most of the time we can't use libc iconv service: - it doesn't round trip for most encoding - it doesn't know about Apple extension " This note appears to be from about 2005. Can I somehow get away with glibc's iconv, or do I, too, need to make a complete copy on my own? Or should I just link afpfs-ng against netatalk's libatalk (which isn't currently shipped as a shared object)? I can see other reasons to use libatalk later... - Alex |