Community
Participate
Working Groups
If I try to create a new workspace or select an existing workspace at the initial workspace selection dialog at startup, it says "Workspace in use or cannot be created, choose a different one." The directory is on an NFS mounted file system (mount options are -2 -T -s -P -b; for performance reasons, I must run NFSv2 at the client, not NFSv3). I'm using Mac OS X 10.4.8 and Eclipse 3.2.1. It does manage to create a lock file, but Eclipse refuses to continue: DING! tack:~/Documents% ls -a workspace/ ./ ../ .metadata/ 3:00pm tack:~/Documents% ls -a workspace/.metadata/ ./ ../ .lock If I create a workspace in a local directory such as /var/tmp/workspace, it works fine. dm
I have the same problem using version 3.3M3.
Are you running the NFS Lock server on the remote system? I'm wondering if that would have an impact. I've been able to run it fine on Mac with AFP on a network share. Mind you, I've had an insane amount of problems with Mac/NFS, so ended up using AFP in the end. Having said all of that, it's better to have your workspace on a local drive, and use a version control system to synchronize changes -- that's how I usually work :-)
I am running the NFS lock server (see below). Mac/NFS does seem pretty cheezy, but using NFSv2 I haven't had any other problems (yet). 1:41pm tack:~/Desktop% rpcinfo -p vision-f1 program vers proto port 100000 2 tcp 111 portmapper 100000 2 udp 111 portmapper 100024 1 udp 10002 status 100024 1 tcp 10002 status 100004 2 udp 834 ypserv 100004 1 udp 834 ypserv 100004 2 tcp 834 ypserv 100004 1 tcp 834 ypserv 600100069 1 udp 835 fypxfrd 600100069 1 tcp 835 fypxfrd 100007 2 udp 893 ypbind 100007 1 udp 893 ypbind 100007 2 tcp 896 ypbind 100007 1 tcp 896 ypbind 100011 1 udp 10005 rquotad 100011 2 udp 10005 rquotad 100011 1 tcp 10005 rquotad 100011 2 tcp 10005 rquotad 100003 2 udp 2049 nfs 100003 3 udp 2049 nfs 100003 4 udp 2049 nfs 100003 2 tcp 2049 nfs 100003 3 tcp 2049 nfs 100003 4 tcp 2049 nfs 100021 1 udp 10001 nlockmgr 100021 3 udp 10001 nlockmgr 100021 4 udp 10001 nlockmgr 100021 1 tcp 10000 nlockmgr 100021 3 tcp 10000 nlockmgr 100021 4 tcp 10000 nlockmgr 100005 1 udp 10004 mountd 100005 1 tcp 10004 mountd 100005 2 udp 10004 mountd 100005 2 tcp 10004 mountd 100005 3 udp 10004 mountd 100005 3 tcp 10004 mountd 100009 1 udp 836 yppasswdd
I am seeing a similar issue on an x86 box running Redhat Enterprise 4, update 4 ( rhel4 ) and trying to use a mounted nfs filesystem for my workspace. I have tried a plethora of options, including a suggested -Dosgi.locking=none to prevent locking, to no avail. I think that this bug should be of fairly high importance, since many developers use nfs workspaces.
I have tried to reproduce to reproduce this on red hat linux machine but have not had any luck. As long as the user has rw access to the file <workspace location>/.metadata/.lock file then I can open a workspace which is located on an NFS mounted drive.
Changing OS from Mac OS to Mac OS X as per bug 185991
We tried eclipse PDT 1.0 on two different Macs and both have same problems. we cannot create a workspace neither on nfs nor on smb. perhaps it could be helpful. creating a symbolic link (ln -s) within the users directory pointing to a network directory works. eclipse won't get an exclusiv lock for network drives.
Isn't this one a dup of bug 44735? And the bug 59780 is proposed as (not real) UI fix?
*** This bug has been marked as a duplicate of bug 44735 ***
