Re: NFS file modes consistency among different operating systems
On Tue, Sep 17, 2013 at 1:28 AM, aurfalien aurfal...@gmail.com wrote: When a file is modified by a user , Whats that users umask? - aurf 755 Thank you very much . Mehmet Erol Sanliturk ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS file modes consistency among different operating systems
On Tue, Sep 17, 2013 at 1:32 AM, aurfalien aurfal...@gmail.com wrote: When a file is modified by a user Also curious whats that users group? - aurf Linux user a: 1000 in group :1000 group n id : 1001 ( member : a ) FreeBSD : user b : 1001 in group 1001 NFS Server : group id : 1000 User a is not able to use files created or modified by user b , and vice versa . Users a and b are not able to use or modify files created or modified by Windows XP user . There is no any restriction for the Windows XP user . Thank you very much . Mehmet Erol Sanliturk ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS file modes consistency among different operating systems
On Sep 16, 2013, at 11:27 PM, Mehmet Erol Sanliturk wrote: On Tue, Sep 17, 2013 at 1:28 AM, aurfalien aurfal...@gmail.com wrote: When a file is modified by a user , Whats that users umask? - aurf 755 Ok, well thats your answer. Only that user can mod the file, every one else has rx privs. I'd highly recommend this book; http://shop.oreilly.com/product/9780596003432.do And book mark this; http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/permissions.html - aurf ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS file modes consistency among different operating systems
From your non MS$ clients, open a shell and type umask, what returns? Sounds like your default umask needs changing is all. I would suggest going with a umask of 775 and ensuring all ppl requiring mod access be group members of what you have settled on. - aurf On Sep 16, 2013, at 8:41 PM, Mehmet Erol Sanliturk wrote: Dear All , I have NFS 3 in FreeBSD 9.1 amd64 . The clients are FreeBSD , Linux , Windows XP through Samba on the same files . The Windows XP is able to access , use and modify files created or modified by any other operating system user . In contrary , FreeBSD and Linux users are NOT able to such sharing because files are created by another user and access mode settings are not changeable due to owner of files . It is very likely that some settings are missing but I do not know which ones . One remedy is to use NFS server in root logged state and change file modes frequently ( An ordinary user in server is NOT permitted to change modes of files created by other users although exported directories owned by such a user ) . How can I solve the following problem : No any client should be able to change file modes set in server All files created by client should inherit modes set in server directory . Thank you very much . Mehmet Erol Sanliturk ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS file modes consistency among different operating systems
On Mon, Sep 16, 2013 at 11:53 PM, aurfalien aurfal...@gmail.com wrote: From your non MS$ clients, open a shell and type umask, what returns? Sounds like your default umask needs changing is all. I would suggest going with a umask of 775 and ensuring all ppl requiring mod access be group members of what you have settled on. - aurf On Sep 16, 2013, at 8:41 PM, Mehmet Erol Sanliturk wrote: Dear All , I have NFS 3 in FreeBSD 9.1 amd64 . The clients are FreeBSD , Linux , Windows XP through Samba on the same files . The Windows XP is able to access , use and modify files created or modified by any other operating system user . In contrary , FreeBSD and Linux users are NOT able to such sharing because files are created by another user and access mode settings are not changeable due to owner of files . It is very likely that some settings are missing but I do not know which ones . One remedy is to use NFS server in root logged state and change file modes frequently ( An ordinary user in server is NOT permitted to change modes of files created by other users although exported directories owned by such a user ) . How can I solve the following problem : No any client should be able to change file modes set in server All files created by client should inherit modes set in server directory . Linux umask : 0002 FreeBSD umask : 0022 Changing client umask to 775 is not solving the problem , because in NFS server , they are setting their own modes without considering existing umask . When a file is modified by a user , the other users in FreeBSD and Linux are not able to access to these files even their umask values are 775 . The Linux user is defined in groups 1000 and 1001 but this is also not permitting access to files modified by other users whether their group is 1000 or 1001 . Thank you very much . Mehmet Erol Sanliturk ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS file modes consistency among different operating systems
When a file is modified by a user , Whats that users umask? - aurf ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS file modes consistency among different operating systems
When a file is modified by a user Also curious whats that users group? - aurf ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Performance: Weirder And Weirder
On Sat, Mar 16, 2013 at 11:49 AM, Tim Daneliuk tun...@tundraware.comwrote: This is really weird. A FreeBSD 9.1 system mounts the following: /dev/ad4s1a989M625M285M69%/ devfs 1.0k1.0k 0B 100%/dev /dev/ad4s1d7.8G 1G6.1G14%/var /dev/ad4s1e 48G9.4G 35G21%/usr /dev/ad4s1f390G127G231G35%/usr1 /dev/ad6s1d902G710G120G86%/usr1/BKU /usr1/something (under ad4s1f) and /usr1/BKU (all of ad6s1d) are exported for NFS mounting on the LAN. I have tested the speeds of these two drives locally doing a 'dd if=/dev/zero '. Their speeds are quite comparable - around 55-60 MB/s so the problem below is not an artifact of a slow drive. The two mounts are imported like this on a Linux Mint 12 machine: machine:/usr1/BKU /BKU nfs rw,soft,intr 0 0 machine:/usr1/shared /shared nfs rw,soft,intr 0 0 Problem: When I write files from the LM12 machines to /BKU the writes are 1/10 the speed of when writing to /shared. Reads are fine in both cases, at near native disk speeds being reported. Someone here suggested I get rid of any symlinks in the mount and I did that to no avail. Incidentally, the only reason I just noticed this is that I upgraded the NIC on the FreeBSD machine and the switch into which it connects to 1000Base because the LM12 machine had a built in 1000Base NIC. I also changed the cables on both machines to ensure they were not the problem. Prior to this, I was bandwidth constrained by the 100Base so I never saw NFS performance as an issue. When I upgraded, I expected faster transfers and when I didn't get them, I started this whole investigation. So ... I'm stumped: - It's not the drive or SATA ports because both drives show comparable performance. - It's not the cables because I can get great throughput on one of the NFS mountpoints. - It's neither NIC for the same reason. Does anyone: A) Have a clue what might be doing this B) Have a suggestion how to track down the problem Thanks, -- --**--** Tim Daneliuk tun...@tundraware.com PGP Key: http://www.tundraware.com/PGP/ With respect to your mount points : /usr1 is spanning TWO different partitions : /dev/ad4s1f390G127G231G35%/usr1 /dev/ad6s1d902G710G120G86%/usr1/BKU because /usr1/BKU is a sub-directory of /usr1 . If you create a new directory , for example /usr2 , and /usr2/BKU , and using this new separate directory for sharing , such as : /dev/ad6s1d902G710G120G86%/usr2/BKU and machine:/usr2/BKU /BKU nfs rw,soft,intr 0 0 will it make difference ? Mehmet Erol Sanliturk ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Performance: Weirder And Weirder
On 03/16/2013 04:20 PM, Mehmet Erol Sanliturk wrote: With respect to your mount points : /usr1 is spanning TWO different partitions : /dev/ad4s1f390G127G231G35%/usr1 /dev/ad6s1d902G710G120G86%/usr1/BKU because /usr1/BKU is a sub-directory of /usr1 . If you create a new directory , for example /usr2 , and /usr2/BKU , and using this new separate directory for sharing , such as : /dev/ad6s1d902G710G120G86%/usr2/BKU and machine:/usr2/BKU /BKU nfs rw,soft,intr 0 0 will it make difference ? Mehmet Erol Sanliturk I just tried this and it made no difference. The same file copied onto the NFS mount on /usr1/shared takes about 20x as long when coppied on to /usr[1|2]/BKU. -- Tim Daneliuk tun...@tundraware.com PGP Key: http://www.tundraware.com/PGP/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Performance: Weirder And Weirder
On Sat, Mar 16, 2013 at 3:07 PM, Tim Daneliuk tun...@tundraware.com wrote: On 03/16/2013 04:20 PM, Mehmet Erol Sanliturk wrote: With respect to your mount points : /usr1 is spanning TWO different partitions : /dev/ad4s1f390G127G231G35%/usr1 /dev/ad6s1d902G710G120G86%/usr1/BKU because /usr1/BKU is a sub-directory of /usr1 . If you create a new directory , for example /usr2 , and /usr2/BKU , and using this new separate directory for sharing , such as : /dev/ad6s1d902G710G120G86%/usr2/BKU and machine:/usr2/BKU /BKU nfs rw,soft,intr 0 0 will it make difference ? Mehmet Erol Sanliturk I just tried this and it made no difference. The same file copied onto the NFS mount on /usr1/shared takes about 20x as long when coppied on to /usr[1|2]/BKU. -- --**--** Tim Daneliuk tun...@tundraware.com PGP Key: http://www.tundraware.com/PGP/ Michael W. Lucas in Absolute FeeBSD , 2nd Edition , ( ISBN : 978-1-59327-151-0 ) , is suggesting the following ( p. 248 ) : In client ( mount , or , fstab ) , use options ( -o tcp , intr , soft , -w=32768 , -r=32768 ) tcp option will request a TCP mount instead of UDP mount , because FreeBSD NFS defaults to running over UDF . This subject may be another check point . Mehmet Erol Sanliturk ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Performance: Weirder And Weirder
just slap an netapp 8.x with an avere flash box in front if you want NFS performance... or isilon. On Sat, Mar 16, 2013 at 5:43 PM, Mehmet Erol Sanliturk m.e.sanlit...@gmail.com wrote: On Sat, Mar 16, 2013 at 3:07 PM, Tim Daneliuk tun...@tundraware.com wrote: On 03/16/2013 04:20 PM, Mehmet Erol Sanliturk wrote: With respect to your mount points : /usr1 is spanning TWO different partitions : /dev/ad4s1f390G127G231G35%/usr1 /dev/ad6s1d902G710G120G86%/usr1/BKU because /usr1/BKU is a sub-directory of /usr1 . If you create a new directory , for example /usr2 , and /usr2/BKU , and using this new separate directory for sharing , such as : /dev/ad6s1d902G710G120G86%/usr2/BKU and machine:/usr2/BKU /BKU nfs rw,soft,intr 0 0 will it make difference ? Mehmet Erol Sanliturk I just tried this and it made no difference. The same file copied onto the NFS mount on /usr1/shared takes about 20x as long when coppied on to /usr[1|2]/BKU. -- --**--** Tim Daneliuk tun...@tundraware.com PGP Key: http://www.tundraware.com/PGP/ Michael W. Lucas in Absolute FeeBSD , 2nd Edition , ( ISBN : 978-1-59327-151-0 ) , is suggesting the following ( p. 248 ) : In client ( mount , or , fstab ) , use options ( -o tcp , intr , soft , -w=32768 , -r=32768 ) tcp option will request a TCP mount instead of UDP mount , because FreeBSD NFS defaults to running over UDF . This subject may be another check point . Mehmet Erol Sanliturk ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Performance: Weirder And Weirder
On 03/16/2013 05:43 PM, Mehmet Erol Sanliturk wrote: Michael W. Lucas in Absolute FeeBSD , 2nd Edition , ( ISBN : 978-1-59327-151-0 ) , is suggesting the following ( p. 248 ) : In client ( mount , or , fstab ) , use options ( -o tcp , intr , soft , -w=32768 , -r=32768 ) tcp option will request a TCP mount instead of UDP mount , because FreeBSD NFS defaults to running over UDF . This subject may be another check point . Another very good suggestion but ... to no avail. Thanks for pointing this out. -- Tim Daneliuk tun...@tundraware.com PGP Key: http://www.tundraware.com/PGP/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Performance: Weirder And Weirder
On Sat, Mar 16, 2013 at 6:46 PM, Tim Daneliuk tun...@tundraware.com wrote: On 03/16/2013 05:43 PM, Mehmet Erol Sanliturk wrote: Michael W. Lucas in Absolute FeeBSD , 2nd Edition , ( ISBN : 978-1-59327-151-0 ) , is suggesting the following ( p. 248 ) : In client ( mount , or , fstab ) , use options ( -o tcp , intr , soft , -w=32768 , -r=32768 ) tcp option will request a TCP mount instead of UDP mount , because FreeBSD NFS defaults to running over UDF . This subject may be another check point . Another very good suggestion but ... to no avail. Thanks for pointing this out. -- --**--** Tim Daneliuk tun...@tundraware.com PGP Key: http://www.tundraware.com/PGP/ I have read messages once more . There is a phrase : Linux Mint 12 machineS ( plural ) . In your descriptions , there is no any information about network setup : Single client , multiple clients , etc . Then , with some assumptions : If there is ONLY ONE client , and all of the tests are performed on this ONLY client , problem may be attributed to FreeBSD server or kind of file(s) in different directories : One of the is encrypted ( requires decryption ) , another is plain file , etc. . If there is MORE than ONE client , problem may be attributed to any one the components of the network ( server , clients , switch , cable , NICs , interfering other software , etc. ) . Assume there is MULTIPLE clients : Take two clients of them : (A) Client 1 : Mount two directories . (B) Client 2 : Mount two directories . Test transmission performance : If they are similar , inspect server settings , directory privileges , etc . , file systems ( one is ZFS , other is UFS2 , etc. ) . All of the hardware may work properly , but if the file reading is not able to feed NIC sufficiently fast , it may show up as degraded performance . Increasing NIC buffer size ( as standard it is around 1000 bytes ) to maximum available , may offset latency of supply of data to NIC . If they are different : Check client specialties : A cable may be CAT5 ( only maximum 100 Mbits transfer . Network cards are adaptive , they try 1 Gbits , if it is not achievable , it reduces to speed to 100 Mbits , even to 10 Mbits ) . In that case either use CAT6 cable or CAT5x ( for 1 Gbit transmission , I do not remember x now ) The cable kind should be written on cable , if it is not written , select a properly labelled cable . Interchange cable tips to clients : If performance interchanges also : Cable or SWITCH port is faulty : Check switch port : It may be a 100 Mbits , be sure that it is also 1 Gbits and working properly . Mehmet Erol Sanliturk ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Performance: Weirder And Weirder
On 03/16/2013 10:15 PM, Mehmet Erol Sanliturk wrote: On Sat, Mar 16, 2013 at 6:46 PM, Tim Daneliuk tun...@tundraware.com mailto:tun...@tundraware.com wrote: On 03/16/2013 05:43 PM, Mehmet Erol Sanliturk wrote: Michael W. Lucas in Absolute FeeBSD , 2nd Edition , ( ISBN : 978-1-59327-151-0 ) , is suggesting the following ( p. 248 ) : In client ( mount , or , fstab ) , use options ( -o tcp , intr , soft , -w=32768 , -r=32768 ) tcp option will request a TCP mount instead of UDP mount , because FreeBSD NFS defaults to running over UDF . This subject may be another check point . Another very good suggestion but ... to no avail. Thanks for pointing this out. -- --__--__ Tim Daneliuk tun...@tundraware.com mailto:tun...@tundraware.com PGP Key: http://www.tundraware.com/PGP/ I have read messages once more . There is a phrase : Linux Mint 12 machineS ( plural ) . In your descriptions , there is no any information about network setup : Single client , multiple clients , etc . Then , with some assumptions : If there is ONLY ONE client , and all of the tests are performed on this ONLY client , problem may be attributed to FreeBSD server or kind of file(s) in different directories : One of the is encrypted ( requires decryption ) , another is plain file , etc. . There is one server - FreeBSD, and one client - LM12. Both have had their cables replaced with new CAT6 wiring. Copying the exact same file to each of the NFS mounts exhibits the problem. Reading from the two NFS mount is fast and as expected, so I do not suspect network issues. The two drives used on the server show similar disk performance locally. The server side exports are identical for both mounts as are the client side mounts. The ONLY difference is that the fast NFS mount has server side permissions of 777 whereas the slow NFS mount has server side permissions of 775. Both are owned by root:wheel. The contents of each filesystem are owned by a user in the wheel group. The one other difference is that all the contents of the slow mount are in a particular user group, and all the ones in the fast mount are in the wheel group. Changing the group ownership of all the stuff in the slow mount to wheel makes no difference. The problem appears to be size related on the slow mount. When I copy, say, a 100MB file to it, performance is just fine. When I copy a 1G file, it's 1/20 the throughput (45MB/sec vs 2MB/sec). This feels like some kind of buffer starvation but the fact that I can run at full speed against another mount point leaves me scratching my head as to just where. It's almost like there's some kind of halting going on during the transfer. Tim Daneliuk tun...@tundraware.com PGP Key: http://www.tundraware.com/PGP/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Performance: Weirder And Weirder
There is one more point to check : From your mount information , in the server , directories are on DIFFERENT drives . Assume one of the drives is very INTELLIGENT to save power . During local reading , due to reading speed , it may not go to SLEEP , but during network access , it may go to sleep due to its exceeded waiting time . If this is the case , please stay away from INTELLIGENT drives in a server : These are designed and produced by very IGNORANT entities . For simple , personal applications , their latency may not be noticed very much , but in a server , they can not be used . Another point may be file sizes . To check effect of file size , into the two different directories copy a large ( for example , 5 GB , or a 4.n GB .iso file ) and transmit these same files from their directories to a single client . If directory structure makes a difference , assuming hardware parts and client does not behave differently to these files , performance difference may be attributed to server side . Mehmet Erol Sanliturk ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS client over private network
Hi, I have some trouble on a mail server running 9.0-RELEASE-p3 Last week I set up a NFS mounted partition containing 1 Tb of IMAP folders the NFS mount is done through a private network link on a dedicated giga ethernet link with the following config : 10.0.0.1/24 -- 10.0.0.2/24 mailhub NFS NetAPP 3210 server I use MBOX format on this server ( Postfix + Dovecot 2 ) There is something somewhere saying don't use mbox format on NFS. It's simply wont work. It may work, but someday you'll face some issue. Time to change for maildir format :) Bon courage. Olivier Since I used this configuration a lock problem occured on some INBOXes here is an example of the problem. Nov 30 23:59:26 mail postfix/local[35280]: 3YCr470N6MzYmp0: to=x...@esiee.fr, orig_to=x...@esiee.fr, relay=local, delay=1527, delays=1470/39/0/19, dsn=4.2.0, status=deferred (cannot update mailbox /var/mail/xxx for user xxx. unable to lock for exclusive access: Resource temporarily unavailable) I don't think Postfix or Dovecot are responsible because when I do not use the private NFS link this does not happen ... I mean when the partition is NFS mounted with public IP addess. Thanks for any info if you have some ... ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS client over private network
On 12/03/2012 10:11 AM, Olivier Nicole wrote: Hi, I have some trouble on a mail server running 9.0-RELEASE-p3 Last week I set up a NFS mounted partition containing 1 Tb of IMAP folders the NFS mount is done through a private network link on a dedicated giga ethernet link with the following config : 10.0.0.1/24 -- 10.0.0.2/24 mailhub NFS NetAPP 3210 server I use MBOX format on this server ( Postfix + Dovecot 2 ) There is something somewhere saying don't use mbox format on NFS. It's simply wont work. It may work, but someday you'll face some issue. Time to change for maildir format :) Bon courage. Olivier A mbox -- maildir convertion on this server will take ~24 hours ... I cannot stop it so long for now ( I have to wait until holidays ... ) Since I used this configuration a lock problem occured on some INBOXes here is an example of the problem. Nov 30 23:59:26 mail postfix/local[35280]: 3YCr470N6MzYmp0: to=x...@esiee.fr, orig_to=x...@esiee.fr, relay=local, delay=1527, delays=1470/39/0/19, dsn=4.2.0, status=deferred (cannot update mailbox /var/mail/xxx for user xxx. unable to lock for exclusive access: Resource temporarily unavailable) I don't think Postfix or Dovecot are responsible because when I do not use the private NFS link this does not happen ... I mean when the partition is NFS mounted with public IP addess. Thanks for any info if you have some ... ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Install
On Thu, 13 Sep 2012 11:29:47 -0400, Gardner Bell wrote: What I'm wanting to do is build/installworld from my workstation to a remote machine but both have different /etc/src.conf and kernel configuration files. Is there a way to define seperate files so I can perform this upgrade without any errors? I assume that you run i386 _or_ amd64 on both systems. It's easy to copy the remote machine's /etc/src.conf to the system you're building on and to _temporarily_ replace the /etc/src.conf of that system. Also copy the kernel configuration file and put it into the correct location (/sys/i386/conf or /sys/amd64/conf). Make sure /usr/obj is empty. Then use the build and install parameter DESTDIR= and pay attention to other upgrading steps as listed in the comment header of /usr/src/Makefile. Also see The FreeBSD Handbook, section 25.7: http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/makeworld.html Rgarding /etc/rc.conf, I'm not aware of specifying a _different_ file name than the default one (i. e., what KERNCONF= does to override GENERIC), so maybe maybe dealing with a symlink in /etc/ would be the least painful way: src.conf - src.conf.local (fits the machine you build on) _or_ - src.conf.remote (fits the machine you build for). You could also create symlinks pointing to their location on the NFS file system (that the remote machine promotes to the build system): src.conf - src.conf.local (fits the machine you build on) _or_ (now mounted via NFS) - /mnt/remotehost/etc/src.conf (fits the machine you build for); and a similar symlink for /sys/i386/conf/REMOTE - /mnt/remotehost/sys/i386/conf/MYKERNEL. I know that looks ugly, but it's the easies solution that currently occurs to my mind. :-) -- Polytropon Magdeburg, Germany Happy FreeBSD user since 4.0 Andra moi ennepe, Mousa, ... ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Install
Hi Gardner Polytropon wrote: On Thu, 13 Sep 2012 11:29:47 -0400, Gardner Bell wrote: What I'm wanting to do is build/installworld from my workstation to a remote machine but both have different /etc/src.conf and kernel configuration files. Is there a way to define seperate files so I can perform this upgrade without any errors? I fairly often do make installs over amd+nfs, a few gotchas to avoid getting caught on: - I saw my link count break in /rescue so du exploded (up by presumably about (137-1 x 4.7M ) (Cant remember why, I just fixed it) - Chflags bit me (maybe I didnt have the right stuff in /etc/exports on target. (I hate chflags. often run chflags -R noschg / ) - If both might be i386, target might 686 eg 586 etc avoid source host having any files lurking in /usr/obj that were built while /etc/make.conf ( included files) had a CFLAGS += -march=i686 Maybe practice on a local host first, where you can reach reset knob. Cheers, Julian -- Julian Stacey, BSD Unix Linux C Sys Eng Consultant, Munich http://berklix.com Reply below not above, like a play script. Indent old text with . Send plain text. Not: HTML, multipart/alternative, base64, quoted-printable. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS within a Jail?!
http://forums.freebsd.org/showthread.php?t=29968highlight=nfsd Found this which I think says it all at the conclusion. March 30th, 2012 Received some information from the FreeBSD mailing list and apparently exporting NFS from an jailed environment is not possible. For those who have managed (by heavy tweaking of sysctl.conf) to export the NFS probably have these concerns: 1) Security may have been compromised on their own jails as a result of tweaks and 2) Even if you manage to export the NFS share under such strained boundary conditions, it may cause problems in some of the application's you would like to use (eg: tinderbox) finally, 3) If you try to use net/unfs3 and succeed to export NFS, this will not have a very fast (ro) transport rate and will have many (rw) speed limitations. My personal conclusion is to wait until the default kernel version of nfs is updated to be jail-friendly before I try using nfs in jails. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS within a Jail?!
http://forums.freebsd.org/showthread.php?t=29968highlight=nfsd Found this which I think says it all at the conclusion. you are truly funny. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS within a Jail?!
Now here is the KEY. No where does it say it has the server side function, only the client side. /usr/ports/net/unfs3 Even if someone proved that it doesn't do NFS server work, i will continue to use it as NFS server, because it works very well ;) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS within a Jail?!
Wojciech Puchar wrote: Now here is the KEY. No where does it say it has the server side function, only the client side. /usr/ports/net/unfs3 Even if someone proved that it doesn't do NFS server work, i will continue to use it as NFS server, because it works very well ;) Since your the expert on unfs3 because you have it working, would you share some technical configuration information with us? Such as What statements do you have in the host:server and remote:client /etc/rc.conf to auto start them at boot time? How do you disable the kernel nfs version so it don't interfere with unfs3? What does your export file look like on both the host:server and remote:client sides. Then about unfs3 performance; how many concurrent remote:clients do you service? Doe's access elapse time get longer as more concurrent remote:clients come online? Do you run unfs3 in a jail on both the host:server and remote:client sides? Are there any sysctl nob settings needed to make unfs3 run in a jail? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS within a Jail?!
Since your the expert on unfs3 because you have it working, would you share some technical configuration information with us? Such as What statements do you have in the host:server and remote:client /etc/rc.conf to auto start them at boot time? in client - as with any NFS, use kernel klient. on server - just run unfsd anywhere, like /etc/rc.local Do you really cannot use any program without rc.d script? ;) How do you disable the kernel nfs version so it don't interfere with unfs3? Just don't enable it in /etc/rc.conf You may have it in kernel. i did, now i don't. both works. What does your export file look like on both the host:server and remote:client sides. as described in manual of unfsd. basics are same, details are not. Then about unfs3 performance; how many concurrent remote:clients do you service? Doe's access elapse time get longer as more concurrent over 60 but not high load clients so performance doesn't matter. That's X terminals booting over NFS. Sometimes i do more - example is net-booting windoze PC to be able to do some recovery OR backup large amount of data to server. There is no practical performance difference in that settings. BUT - do make configure, then search where fsync is called and comment it out. Right - not conformant, but the performance difference on writes are enormous. Unless you do such a stupid things like running database servers over NFS, you don't need this conformance. Just do it. Do you run unfs3 in a jail on both the host:server and remote:client sides? Not now. but tried. unfs doesn't need ANY special kernel calls. It runs just like any program using UDP/TCP communication. jail/no jail doesn't make a difference. As with most programs. In kernel point of view it is just a program that use TCP/IP stack and open/read/write/readdir/etc. Nothing else. Are there any sysctl nob settings needed to make unfs3 run in a jail? as above. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS within a Jail?!
Many thanks in advance. Quick answer is No, NFS only runs on the host system. but user space nfsd works. in ports - unfsd ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS within a Jail?!
Really? Is that stable enough to serve files for months without disruption? Wojciech Puchar woj...@wojtek.tensor.gdynia.pl wrote: Many thanks in advance. Quick answer is No, NFS only runs on the host system. but user space nfsd works. in ports - unfsd -- Sent from my Android phone with K-9 Mail. Please excuse my brevity. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS within a Jail?!
Wojciech Puchar wrote: Many thanks in advance. Quick answer is No, NFS only runs on the host system. but user space nfsd works. in ports - unfsd Close but no cigar. In the ports system it's named unfs3 and described as UNFS3 is a user-space implementation of the NFSv3 server specification. It provides a daemon for the MOUNT and NFS protocols, which are used by NFS clients for accessing files on the server. http://unfs3.sourceforge.net/ Now here is the KEY. No where does it say it has the server side function, only the client side. But it does have 18k downloads even though it's labeled as beta version. So the question back to Wojciech Puchar is; are you running this unfs3 in a client jail on one pc and the server side in a jail on the host? Put another way can you confirm from experience that this unfs3 port has both client and service side support and that it does work when installed in a client jail and host server jail? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS within a Jail?!
On Fri, Aug 10, 2012 at 08:36:04AM -0400, Fbsd8 typed: Wojciech Puchar wrote: Many thanks in advance. Quick answer is No, NFS only runs on the host system. but user space nfsd works. in ports - unfsd Close but no cigar. In the ports system it's named unfs3 and described as UNFS3 is a user-space implementation of the NFSv3 server specification. ^^ It provides a daemon for the MOUNT and NFS protocols, which are used by ^^ ^^^ NFS clients for accessing files on the server. http://unfs3.sourceforge.net/ Now here is the KEY. No where does it say it has the server side function, only the client side. It sais so very clearly. Ruben ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS within a Jail?!
Ruben de Groot wrote: On Fri, Aug 10, 2012 at 08:36:04AM -0400, Fbsd8 typed: Wojciech Puchar wrote: Many thanks in advance. Quick answer is No, NFS only runs on the host system. but user space nfsd works. in ports - unfsd Close but no cigar. In the ports system it's named unfs3 and described as UNFS3 is a user-space implementation of the NFSv3 server specification. ^^ It provides a daemon for the MOUNT and NFS protocols, which are used by ^^ ^^^ NFS clients for accessing files on the server. http://unfs3.sourceforge.net/ Now here is the KEY. No where does it say it has the server side function, only the client side. It sais so very clearly. Ruben What you say: Just 2 words further on in that sentence used by NFS clients. Read as unfs3 is run as client to access kernel nfs on host. No where in any documentation on unfs3 does it ever say unfs3 has to be run on both client and service side. That sentence infers that unfs3 is only run on the client side. Now I do concede that the writer of that sentence may not be a native English speaker and as such fails to express fully the intent of what he was trying to say. Maybe unfs3 really has to be used on both the host server side and clients side for it to work. Or this may just be a case of the author being to close to the trees to see the forest. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS within a Jail?!
On Wed, Aug 8, 2012 at 1:57 PM, Fbsd8 fb...@a1poweruser.com wrote: Dealing with this has been SOP practice in jails since their inception. See man 8 jail. The best way to run the NFS server is from the jail. Running it host side is the hard part. http://www.freebsd.org/cgi/**query-pr.cgi?pr=133265http://www.freebsd.org/cgi/query-pr.cgi?pr=133265 The jail code maintainer says NFS server/client will not work jailed. So since you say this is SOP (standard operation procedure) then why is there no documentation available on how to do it? All the Google hits for NFS running from Freebsd jail end with no one got it to work. Have you done this? Do you have a procedure to post or know of a posted procedure giving step-by-step sequence to get NFS running in a jail with or without VIMAGE/VNET for Release 8.x or 9.x versions? That PR is about mounting a fs in a jail, specifically one proved by NFS. What does that have to do with the OP's question? It's quite clear you didn't read the full thing. Still doesn't change the FACT it's experimental! Which is your sole reason for poo=pooing it? Are you talking about the arbitrary line between experimental and production? I wonder how a piece of functionality transitions from experimental to production...is it possible we get there by promoting mindshare of the new piece instead of FUD? -- Adam Vande More ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS within a Jail?!
On Fri, Aug 10, 2012 at 10:42:45AM -0400, Fbsd8 typed: Ruben de Groot wrote: On Fri, Aug 10, 2012 at 08:36:04AM -0400, Fbsd8 typed: Wojciech Puchar wrote: Many thanks in advance. Quick answer is No, NFS only runs on the host system. but user space nfsd works. in ports - unfsd Close but no cigar. In the ports system it's named unfs3 and described as UNFS3 is a user-space implementation of the NFSv3 server specification. ^^ It provides a daemon for the MOUNT and NFS protocols, which are used by ^^ ^^^ NFS clients for accessing files on the server. http://unfs3.sourceforge.net/ Now here is the KEY. No where does it say it has the server side function, only the client side. It sais so very clearly. Ruben What you say: Just 2 words further on in that sentence used by NFS clients. Read as unfs3 is run as client to access kernel nfs on host. No you read wrong. It is a userspace daemon that provides nfs service. It is a daemon. Not a client. No where in any documentation on unfs3 does it ever say unfs3 has to be run on both client and service side. Because it is not a client. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS within a Jail?!
blackfriar wrote: Hi everybody! I'm wondering if it's possible to run in a clear fashion an NFS server within a jail on FreeBSD 9.0. I'm having some issues that make me think this is not supposed to work. I've googled it but I couldn't find much especially on releases prior 5!! A quick tip would be great I don't really wanna waste hours on this not very relevant issue. Many thanks in advance. Quick answer is No, NFS only runs on the host system. Long answer is; NFS requires rpcbind to function. rpcbind is dependent on a network stack. Jails do not have their own network stack, they use the hosts network stack. There is some experimental software to give each jail its own network stack but I sure would not deploy a production system based on this. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS within a Jail?!
Clear! Thanks a lot. On Wed, 2012-08-08 at 10:02 -0400, Fbsd8 wrote: blackfriar wrote: Hi everybody! I'm wondering if it's possible to run in a clear fashion an NFS server within a jail on FreeBSD 9.0. I'm having some issues that make me think this is not supposed to work. I've googled it but I couldn't find much especially on releases prior 5!! A quick tip would be great I don't really wanna waste hours on this not very relevant issue. Many thanks in advance. Quick answer is No, NFS only runs on the host system. Long answer is; NFS requires rpcbind to function. rpcbind is dependent on a network stack. Jails do not have their own network stack, they use the hosts network stack. There is some experimental software to give each jail its own network stack but I sure would not deploy a production system based on this. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org signature.asc Description: This is a digitally signed message part
Re: NFS within a Jail?!
On Wed, Aug 8, 2012 at 9:02 AM, Fbsd8 fb...@a1poweruser.com wrote: Long answer is; NFS requires rpcbind to function. rpcbind is dependent on a network stack. Jails do not have their own network stack, they use the hosts network stack. Dealing with this has been SOP practice in jails since their inception. See man 8 jail. The best way to run the NFS server is from the jail. Running it host side is the hard part. There is some experimental software to give each jail its own network stack but I sure would not deploy a production system based on this. There are a number of people who have reached the opposite decision concerning VIMAGE/VNET enabled jails. They are much easier to work with and provide nice capabilities. -- Adam Vande More ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS within a Jail?!
On Wed, 08 Aug 2012 11:55:51 -0500, Adam Vande More amvandem...@gmail.com wrote: There are a number of people who have reached the opposite decision concerning VIMAGE/VNET enabled jails. They are much easier to work with and provide nice capabilities. I tried it on 9.0-RELEASE and was able to cause kernel panics quite easily, so I've avoided it. I expect things to be worked out by the end of the 9.x train and/or 10.0. They certainly do provide some nice capabilities, though. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS within a Jail?!
Adam Vande More wrote: On Wed, Aug 8, 2012 at 9:02 AM, Fbsd8 fb...@a1poweruser.com wrote: Long answer is; NFS requires rpcbind to function. rpcbind is dependent on a network stack. Jails do not have their own network stack, they use the hosts network stack. Dealing with this has been SOP practice in jails since their inception. See man 8 jail. The best way to run the NFS server is from the jail. Running it host side is the hard part. http://www.freebsd.org/cgi/query-pr.cgi?pr=133265 The jail code maintainer says NFS server/client will not work jailed. So since you say this is SOP (standard operation procedure) then why is there no documentation available on how to do it? All the Google hits for NFS running from Freebsd jail end with no one got it to work. Have you done this? Do you have a procedure to post or know of a posted procedure giving step-by-step sequence to get NFS running in a jail with or without VIMAGE/VNET for Release 8.x or 9.x versions? There is some experimental software to give each jail its own network stack but I sure would not deploy a production system based on this. There are a number of people who have reached the opposite decision concerning VIMAGE/VNET enabled jails. They are much easier to work with and provide nice capabilities. Still doesn't change the FACT it's experimental! ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS mount error: RPCPROG_MNT: RPC: Authentication error; why = Client credential too weak
On 07/06/2012 06:03 PM, Walter Hurry wrote: I am running FreeBSD 9.0-RELEASE (64 bit), with a VirtualBox VM also running the same. On the host I am running NFS server: $ showmount -e Exports list on localhost: /usr/home Everyone But when I try to mount is on the client (the VM guest) I get this: # mount xx:/usr/home /mnt [tcp] xx:/usr/home: RPCPROG_MNT: RPC: Authentication error; why = Client credential too weak # Hi Walter Are you root when mounting on the client? From looking at your prompt # I think you are, but I ask just to make sure. You can also take a look at http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/network-nfs.html in the handbook On the server, in /var/log/messages I see this: mountd[29140]: mount request from nnn.nnn.nnn.nnn from unprivileged port So I infer that the 'unprivileged port' bit is the problem. Further information - on the client: $ rpcinfo xx program version netid addressserviceowner 104tcp 0.0.0.0.0.111 rpcbindsuperuser 103tcp 0.0.0.0.0.111 rpcbindsuperuser 102tcp 0.0.0.0.0.111 rpcbindsuperuser 104udp 0.0.0.0.0.111 rpcbindsuperuser 103udp 0.0.0.0.0.111 rpcbindsuperuser 102udp 0.0.0.0.0.111 rpcbindsuperuser 104tcp6 ::.0.111 rpcbindsuperuser 103tcp6 ::.0.111 rpcbindsuperuser 104udp6 ::.0.111 rpcbindsuperuser 103udp6 ::.0.111 rpcbindsuperuser 104local /var/run/rpcbind.sock rpcbindsuperuser 103local /var/run/rpcbind.sock rpcbindsuperuser 102local /var/run/rpcbind.sock rpcbindsuperuser 132udp 0.0.0.0.8.1nfssuperuser 133udp 0.0.0.0.8.1nfssuperuser 132udp6 ::.8.1 nfssuperuser 133udp6 ::.8.1 nfssuperuser 132tcp 0.0.0.0.8.1nfssuperuser 133tcp 0.0.0.0.8.1nfssuperuser 132tcp6 ::.8.1 nfssuperuser 133tcp6 ::.8.1 nfssuperuser 151udp6 ::.2.94mountd superuser 153udp6 ::.2.94mountd superuser 151tcp6 ::.2.94mountd superuser 153tcp6 ::.2.94mountd superuser 151udp 0.0.0.0.2.94 mountd superuser 153udp 0.0.0.0.2.94 mountd superuser 151tcp 0.0.0.0.2.94 mountd superuser 153tcp 0.0.0.0.2.94 mountd superuser $ What am I doing wrong? I am new to NFS. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org Disclaimer: http://www.ose.nl/email ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS mount error: RPCPROG_MNT: RPC: Authentication error; why = Client credential too weak
On Fri, 06 Jul 2012 12:42:02 -0400, kpneal wrote: On Fri, Jul 06, 2012 at 04:03:27PM +, Walter Hurry wrote: I am running FreeBSD 9.0-RELEASE (64 bit), with a VirtualBox VM also running the same. On the host I am running NFS server: $ showmount -e Exports list on localhost: /usr/home Everyone But when I try to mount is on the client (the VM guest) I get this: # mount xx:/usr/home /mnt [tcp] xx:/usr/home: RPCPROG_MNT: RPC: Authentication error; why = Client credential too weak # On the server, in /var/log/messages I see this: mountd[29140]: mount request from nnn.nnn.nnn.nnn from unprivileged port So I infer that the 'unprivileged port' bit is the problem. That's odd. At 8.2 the documentation (man page) on mount_nfs says that reserved ports are the default. I'd be surprised if 9.0 was different. *shrug* Try running mountd with the -n option. If I understand the man page then it allows mountd to accept mounts from port numbers less than 1024. Note that the mountd protocol is distinct from the NFS protocol and so rpcinfo can't really tell you anything about mountd. Thanks. Yes, the mount worked fine on the client when the server mountd was started with the -n option. That leads me to two more questions: Why would mount_nfs be using an unprivileged port by default? As far as I can see from man mount_nfs the only relevant option would be (section of manpage reformatted for convenience): port=⟨port_number⟩ Use specified port number for NFS requests. The default is to query the portmapper for the NFS port. I'm afraid that due to my lack of knowledge in this area, that doesn't mean a lot to me. Are there security implications in using an unprivileged port? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS mount error: RPCPROG_MNT: RPC: Authentication error; why = Client credential too weak
On Fri, 06 Jul 2012 18:55:27 +0200, Bas Smeelen wrote: Are you root when mounting on the client? From looking at your prompt # I think you are, but I ask just to make sure. You can also take a look at http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/network- nfs.html in the handbook Thanks for the reply. Yes, I'm running as root on the client when I try the mount. It was the handbook I was following in my attempt to set up NFS. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS mount error: RPCPROG_MNT: RPC: Authentication error; why = Client credential too weak
On 07/06/2012 07:25 PM, Walter Hurry wrote: On Fri, 06 Jul 2012 18:55:27 +0200, Bas Smeelen wrote: Are you root when mounting on the client? From looking at your prompt # I think you are, but I ask just to make sure. You can also take a look at http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/network- nfs.html in the handbook Thanks for the reply. Yes, I'm running as root on the client when I try the mount. It was the handbook I was following in my attempt to set up NFS. OK. With -n (allow from non root users) for mountd the mount succeeds although without it doesn't but you are root on the client. The nfs server is use is still 7.4 and I cannot find a difference in the man pages of 7 and 9 mountd and mount_nfs regarding to this issue. In regard to the security implications, I think that we don't want mounts from trusted clients by a non root user who cannot bind to privileged ports, thus deny unprivileged ports. Disclaimer: http://www.ose.nl/email ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS locking and linux NFS server
On 03/25/12 23:59, Christoph Egger wrote: Hi all! I have a Linux Host (2.6.32 kernel, Debian stable) providong NFS shares. Locking files on that share works fine for linux clients [0] while it fails on a freebsd 9.0-STABLE system. The interwebs indicate there have been problems witha buggy linux implementation back in 2006 but no more hits for that problem in recent times so I assume it's fixed? root@freebsd /mnt/ 11:27 0 # kldstat -v | grep nfs 341 nfscommon 386 nfslockd 344 nfsd 385 nfssvc 342 nfs 343 nfscl 384 nfslock root@freebsd /mnt/ 11:28 0 # flock test -c ls flock: test: Operation not supported root@freebsd /mnt/ 11:31 0 # mount | grep nfs 10.70.255.8:/home/ on /mnt/ (nfs) This may or may not be helpful, but I can't think of anything else at this time: what version NFS on both sides? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS locking and linux NFS server
Hi! Da Rock freebsd-questi...@herveybayaustralia.com.au writes: On 03/25/12 23:59, Christoph Egger wrote: Hi all! I have a Linux Host (2.6.32 kernel, Debian stable) providong NFS shares. Locking files on that share works fine for linux clients [0] while it fails on a freebsd 9.0-STABLE system. The interwebs indicate there have been problems witha buggy linux implementation back in 2006 but no more hits for that problem in recent times so I assume it's fixed? root@freebsd /mnt/ 11:27 0 # kldstat -v | grep nfs 341 nfscommon 386 nfslockd 344 nfsd 385 nfssvc 342 nfs 343 nfscl 384 nfslock root@freebsd /mnt/ 11:28 0 # flock test -c ls flock: test: Operation not supported root@freebsd /mnt/ 11:31 0 # mount | grep nfs 10.70.255.8:/home/ on /mnt/ (nfs) This may or may not be helpful, but I can't think of anything else at this time: what version NFS on both sides? NFSv3 on both sides Regards Christoph -- 9FED 5C6C E206 B70A 5857 70CA 9655 22B9 D49A E731 Debian Developer | Lisp Hacker | CaCert Assurer ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS locking and linux NFS server
In the last episode (Mar 25), Christoph Egger said: Hi all! I have a Linux Host (2.6.32 kernel, Debian stable) providong NFS shares. Locking files on that share works fine for linux clients [0] while it fails on a freebsd 9.0-STABLE system. The interwebs indicate there have been problems witha buggy linux implementation back in 2006 but no more hits for that problem in recent times so I assume it's fixed? root@freebsd /mnt/ 11:27 0 # kldstat -v | grep nfs 341 nfscommon 386 nfslockd 344 nfsd 385 nfssvc 342 nfs 343 nfscl 384 nfslock root@freebsd /mnt/ 11:28 0 # flock test -c ls flock: test: Operation not supported root@freebsd /mnt/ 11:31 0 # mount | grep nfs 10.70.255.8:/home/ on /mnt/ (nfs) Are you running statd and lockd (in rc.conf, rpc_statd_enable=YES and rpc_lockd_enable=YES)? Make sure that rpcinfo localhost and rpcinfo otherhost both show nlockmgr and status services. -- Dan Nelson dnel...@allantgroup.com ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS locking and linux NFS server
Hi! Dan Nelson dnel...@allantgroup.com writes: Are you running statd and lockd (in rc.conf, rpc_statd_enable=YES and rpc_lockd_enable=YES)? Make sure that rpcinfo localhost and rpcinfo otherhost both show nlockmgr and status services. it was missing nfs_client_enable=YES Thanks everyone for the answers! Regards Christoph -- 9FED 5C6C E206 B70A 5857 70CA 9655 22B9 D49A E731 Debian Developer | Lisp Hacker | CaCert Assurer ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: nfs client speed lower than expected.
iperf [ ID] Interval Transfer Bandwidth [ 3] 0.0-60.2 sec 6.22 GBytes 887 Mbits/sec transfers in via ssh are nice and nifty too. Vince On 02/11/2011 23:55, Gary Gatten wrote: Is the interface really at 1Gb? Have you tested with iperf, ftp, or anything other than nfs? -Original Message- From: owner-freebsd-questi...@freebsd.org [mailto:owner-freebsd-questi...@freebsd.org] On Behalf Of Vincent Hoffman Sent: Wednesday, November 02, 2011 6:52 PM To: FreeBSD-Questions@freebsd.org Subject: nfs client speed lower than expected. Hi all, What kind of speed should I be expecting over an NFS mount from a linux box using a gig interface (igb)? I'm seeing linux clients getting approx 2 or 3 times the throughput rsyncing files from a linux nfs server that i get from a 8-stable FreeBSD client. representative results 7.26MB/s - Freebsd client 21.10MB/s liunx client I've tried a variety of files to try and take caching out of the equation, I've tweaked a few sysctls after much googling kern.ipc.maxsockbuf=400 net.inet.tcp.sendbuf_max=16777216 net.inet.tcp.recvbuf_max=16777216 net.inet.tcp.sendbuf_inc=16384 net.inet.tcp.recvbuf_inc=524288 net.inet.udp.recvspace=65535 net.local.stream.recvspace=65535 net.local.stream.sendspace=65535 with no obvious improvement. freebsd mount options ro,noatime,noexec,nosuid,udp,nfsv3,rsize=1024,wsize=1024,bg,hard,intr,timeout=4,retrans=4 linux mount options _netdev,ro,noatime,nodev,noexec,nosuid,proto=udp,vers=3,rsize=1k,wsize=1k,bg,hard,intr,timeo=4,retrans=4 I have seen that using the linux server as an nfs client to write to the NFS server on the freebsd box gives similar performance to a linux client pulling from the linux server so I'm guessing its something to do with the freebsd nfs client? Any suggestions/clues welcome. Thanks, Vince ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org font size=1 div style='border:none;border-bottom:double windowtext 2.25pt;padding:0in 0in 1.0pt 0in' /div This email is intended to be reviewed by only the intended recipient and may contain information that is privileged and/or confidential. If you are not the intended recipient, you are hereby notified that any review, use, dissemination, disclosure or copying of this email and its attachments, if any, is strictly prohibited. If you have received this email in error, please immediately notify the sender by return email and delete this email from your system. /font ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: nfs client speed lower than expected.
I'll give it a try when i get a moment, These servers are all on gigabit LAN (sadly 1500 mtu until I can get the network guy to schedule an outage to reboot the switches and enable jumbo frames,) same subnet so i would expect UDP to have similar or better performance. Vince On 02/11/2011 23:53, Michael Sierchio wrote: Mount via tcp. On Wed, Nov 2, 2011 at 4:51 PM, Vincent Hoffman vi...@unsane.co.uk wrote: Hi all, What kind of speed should I be expecting over an NFS mount from a linux box using a gig interface (igb)? I'm seeing linux clients getting approx 2 or 3 times the throughput rsyncing files from a linux nfs server that i get from a 8-stable FreeBSD client. representative results 7.26MB/s - Freebsd client 21.10MB/s liunx client I've tried a variety of files to try and take caching out of the equation, I've tweaked a few sysctls after much googling kern.ipc.maxsockbuf=400 net.inet.tcp.sendbuf_max=16777216 net.inet.tcp.recvbuf_max=16777216 net.inet.tcp.sendbuf_inc=16384 net.inet.tcp.recvbuf_inc=524288 net.inet.udp.recvspace=65535 net.local.stream.recvspace=65535 net.local.stream.sendspace=65535 with no obvious improvement. freebsd mount options ro,noatime,noexec,nosuid,udp,nfsv3,rsize=1024,wsize=1024,bg,hard,intr,timeout=4,retrans=4 linux mount options _netdev,ro,noatime,nodev,noexec,nosuid,proto=udp,vers=3,rsize=1k,wsize=1k,bg,hard,intr,timeo=4,retrans=4 I have seen that using the linux server as an nfs client to write to the NFS server on the freebsd box gives similar performance to a linux client pulling from the linux server so I'm guessing its something to do with the freebsd nfs client? Any suggestions/clues welcome. Thanks, Vince ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: nfs client speed lower than expected.
Mount via tcp. On Wed, Nov 2, 2011 at 4:51 PM, Vincent Hoffman vi...@unsane.co.uk wrote: Hi all, Â Â Â Â What kind of speed should I be expecting over an NFS mount from a linux box using a gig interface (igb)? I'm seeing linux clients getting approx 2 or 3 times the throughput rsyncing files from a linux nfs server that i get from a 8-stable FreeBSD client. representative results 7.26MB/s - Freebsd client 21.10MB/s liunx client I've tried a variety of files to try and take caching out of the equation, I've tweaked a few sysctls after much googling kern.ipc.maxsockbuf=400 net.inet.tcp.sendbuf_max=16777216 net.inet.tcp.recvbuf_max=16777216 net.inet.tcp.sendbuf_inc=16384 net.inet.tcp.recvbuf_inc=524288 net.inet.udp.recvspace=65535 net.local.stream.recvspace=65535 net.local.stream.sendspace=65535 with no obvious improvement. freebsd mount options ro,noatime,noexec,nosuid,udp,nfsv3,rsize=1024,wsize=1024,bg,hard,intr,timeout=4,retrans=4 linux mount options _netdev,ro,noatime,nodev,noexec,nosuid,proto=udp,vers=3,rsize=1k,wsize=1k,bg,hard,intr,timeo=4,retrans=4 I have seen that using the linux server as an nfs client to write to the NFS server on the freebsd box gives similar performance to a linux client pulling from the linux server so I'm guessing its something to do with the freebsd nfs client? Any suggestions/clues welcome. Thanks, Vince ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
RE: nfs client speed lower than expected.
Is the interface really at 1Gb? Have you tested with iperf, ftp, or anything other than nfs? -Original Message- From: owner-freebsd-questi...@freebsd.org [mailto:owner-freebsd-questi...@freebsd.org] On Behalf Of Vincent Hoffman Sent: Wednesday, November 02, 2011 6:52 PM To: FreeBSD-Questions@freebsd.org Subject: nfs client speed lower than expected. Hi all, What kind of speed should I be expecting over an NFS mount from a linux box using a gig interface (igb)? I'm seeing linux clients getting approx 2 or 3 times the throughput rsyncing files from a linux nfs server that i get from a 8-stable FreeBSD client. representative results 7.26MB/s - Freebsd client 21.10MB/s liunx client I've tried a variety of files to try and take caching out of the equation, I've tweaked a few sysctls after much googling kern.ipc.maxsockbuf=400 net.inet.tcp.sendbuf_max=16777216 net.inet.tcp.recvbuf_max=16777216 net.inet.tcp.sendbuf_inc=16384 net.inet.tcp.recvbuf_inc=524288 net.inet.udp.recvspace=65535 net.local.stream.recvspace=65535 net.local.stream.sendspace=65535 with no obvious improvement. freebsd mount options ro,noatime,noexec,nosuid,udp,nfsv3,rsize=1024,wsize=1024,bg,hard,intr,timeout=4,retrans=4 linux mount options _netdev,ro,noatime,nodev,noexec,nosuid,proto=udp,vers=3,rsize=1k,wsize=1k,bg,hard,intr,timeo=4,retrans=4 I have seen that using the linux server as an nfs client to write to the NFS server on the freebsd box gives similar performance to a linux client pulling from the linux server so I'm guessing its something to do with the freebsd nfs client? Any suggestions/clues welcome. Thanks, Vince ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org font size=1 div style='border:none;border-bottom:double windowtext 2.25pt;padding:0in 0in 1.0pt 0in' /div This email is intended to be reviewed by only the intended recipient and may contain information that is privileged and/or confidential. If you are not the intended recipient, you are hereby notified that any review, use, dissemination, disclosure or copying of this email and its attachments, if any, is strictly prohibited. If you have received this email in error, please immediately notify the sender by return email and delete this email from your system. /font ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS zfs serveur (hardware question)
On Sat, Jun 25, 2011 at 5:07 AM, Peter Toth free...@snap.net.nz wrote: There is still a way to increase NFS performance in 9.0 (without a ZIL SSD) by setting zfs property sync=disabled, which will disable synchronous writes - comes with some risks, research it before switching it off. Also, this will only disable sync for the ZFS filesystem not for the whole pool. Thanks, I'll look into that. I do appreciate that ZFS tries to be more careful about sync writes than most filesystems. But I also have users who expect tar xvf to complete in a reasonable amount of time, and having the ZIL enabled reduces file creation performance by a factor of ten. ;) ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS zfs serveur (hardware question)
On 06/24/11 10:17, David Brodbeck wrote: On Wed, Jun 22, 2011 at 6:26 PM, Damien Fleuriot m...@my.gd wrote: As a rule of thumb and for a serious server, I would recommend 1 SSD as dedicated cache and 2 SSD for a mirrored ZIL (you don't want to lose this data). However I think ppl posted about running intro trouble when using both ZIL and cache disks, so I suggest you only get the ZIL. Definitely get the ZIL device. NFS performance will be almost intolerable without it. It used to be you could work around this, at cost of an increased risk of data loss if the server crashed, by turning off the ZIL; but as of 9.0 this is no longer allowed, so a ZIL device is pretty much mandatory. I'm looking at ways to add one to one of my machines for this reason. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org There is still a way to increase NFS performance in 9.0 (without a ZIL SSD) by setting zfs property sync=disabled, which will disable synchronous writes - comes with some risks, research it before switching it off. Also, this will only disable sync for the ZFS filesystem not for the whole pool. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS zfs serveur (hardware question)
On Wed, June 22, 2011 9:26 pm, Damien Fleuriot wrote: You will lose your main ZFS pool if you lose: - more than 1 of your full ZFS pools or - your ZIL (need confirmation on that) From my reading, on the ZIL: Under 8.2, true. If you have patched your ZFS install, or are running -CURRENT, you can lose your ZIL, I think. (The ability is in zpool version 19.) The 'I think' is because that version allows *removal* of the ZIL device. Which should be the same as a loss of the device, but... Daniel T. Staal --- This email copyright the author. Unless otherwise noted, you are expressly allowed to retransmit, quote, or otherwise use the contents for non-commercial purposes. This copyright will expire 5 years after the author's death, or in 30 years, whichever is longer, unless such a period is in excess of local copyright law. --- ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS zfs serveur (hardware question)
On Wed, Jun 22, 2011 at 6:26 PM, Damien Fleuriot m...@my.gd wrote: As a rule of thumb and for a serious server, I would recommend 1 SSD as dedicated cache and 2 SSD for a mirrored ZIL (you don't want to lose this data). However I think ppl posted about running intro trouble when using both ZIL and cache disks, so I suggest you only get the ZIL. Definitely get the ZIL device. NFS performance will be almost intolerable without it. It used to be you could work around this, at cost of an increased risk of data loss if the server crashed, by turning off the ZIL; but as of 9.0 this is no longer allowed, so a ZIL device is pretty much mandatory. I'm looking at ways to add one to one of my machines for this reason. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS zfs serveur (hardware question)
On Wed, 22 Jun 2011 08:41:58 -0500, Michel Le Cocq miconof80.l...@gmail.com wrote: speedy disk : Sas 15K : to limit IO Wait The more spindles the better. Get more disks if possible. Regards, Mark ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: nfs error: No route to host when starting apache ...
On Fri, 1 Apr 2011, Rick Macklem wrote: Since rpc.lockd and rpc.statd expect to be able to do IP broadcast (same goes for rpcbind), I suspect that might be a problem w.r.t. jails, although I know nothing about how jails work? Oh, and you can use the nolock mount option to avoid use of rpc.lockd and rpc.statd. based on the mount_nfs man page, as well as trying it just in case, this option no longer appears to be availalble in the 7.x nfs code ... :( Oops, sorry. The option is called nolockd. rick ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: nfs error: No route to host when starting apache ...
On Fri, 1 Apr 2011, Rick Macklem wrote: Since rpc.lockd and rpc.statd expect to be able to do IP broadcast (same goes for rpcbind), I suspect that might be a problem w.r.t. jails, although I know nothing about how jails work? Oh, and you can use the nolock mount option to avoid use of rpc.lockd and rpc.statd. based on the mount_nfs man page, as well as trying it just in case, this option no longer appears to be availalble in the 7.x nfs code ... :( ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: nfs error: No route to host when starting apache ...
I've succeedig in getting a bit further ... by the time I got to the bottom of my original, I started to think in terms of rpc more, and had overlooked lookign at thte rpcbind man page, which *does* have a -h option ... setting that fixes things perfectly *almost* ... The last issue I seem to be hitting *might* be a 6.x NFS client against a 7.x server issue ... ? Postfix generates: postfix/showq[65261]: fatal: select lock: Permission denied The only post I found about this was: http://lists.freebsd.org/pipermail/freebsd-questions/2010-April/215284.html But there didn't appear to be any responses ... so either all responses were private to Robert, or ... ? This is my last 6.x box, so it is not overly critical, but would be nice if I could get it to work properly ... On Fri, 1 Apr 2011, Marc G. Fournier wrote: I just setup an nfs mount between two servers ... ServerA, nfsd on 192.168.1.8 ServerB, nfs client on 192.168.1.7 I have a jail, ServerC, running on 192.168.1.7 ... most operations appear to work, but it looks like 'special files' of a sort aren't working, for when I try and startup Apache, I get: [Fri Apr 01 19:42:02 2011] [emerg] (65)No route to host: couldn't grab the accept mutex When I try and do a 'newaliases', I get: # newaliases postalias: fatal: lock /etc/aliases.db: No route to host Yet, for instance, both MySQL and PostgreSQL are running without any issues ... So, the mount is there, it is readable, it is working ... I can ssh into the jail, I can create files, etc ... I do have rpc.lockd and rpc.statd running on both client / server sides ... I'm not seeing anything in eithr the man page for mount_nfs *or* nfsd that might account / corect for something like this, but since I'm not sure what this is exactly, not sure exactl what I should be looking for :( Note that this behaviour happens at the *physical* server level as well, having tested with using postalias to generate the same 'lock' issue above ... Now, I do have mountd/nfsd started iwth the -h to bind them to 192.168.1.8 ... *but*, the servers themselves, although on same switch do have different default gateways ... I'm not seeing anything within the man page for, say, rpc.statd/rpc.lockd that allows me to bind it to the 192.168.1.0/24 IP, so is it binding to my public IP instead of my private? So nfsd / mount_nfs can talk find, as they go thorugh 192.168.1.0/24 as desired, but rpc.statd/rpc.lockd are the public IPs and not able to talk to each other? Thx ... ___ freebsd-...@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to freebsd-net-unsubscr...@freebsd.org Marc G. FournierHub.Org Hosting Solutions S.A. scra...@hub.org http://www.hub.org Yahoo:yscrappySkype: hub.orgICQ:7615664MSN:scra...@hub.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: nfs error: No route to host when starting apache ...
I just setup an nfs mount between two servers ... ServerA, nfsd on 192.168.1.8 ServerB, nfs client on 192.168.1.7 I have a jail, ServerC, running on 192.168.1.7 ... most operations appear to work, but it looks like 'special files' of a sort aren't working, for when I try and startup Apache, I get: [Fri Apr 01 19:42:02 2011] [emerg] (65)No route to host: couldn't grab the accept mutex When I try and do a 'newaliases', I get: # newaliases postalias: fatal: lock /etc/aliases.db: No route to host Yet, for instance, both MySQL and PostgreSQL are running without any issues ... So, the mount is there, it is readable, it is working ... I can ssh into the jail, I can create files, etc ... I do have rpc.lockd and rpc.statd running on both client / server sides ... Since rpc.lockd and rpc.statd expect to be able to do IP broadcast (same goes for rpcbind), I suspect that might be a problem w.r.t. jails, although I know nothing about how jails work? I'm not seeing anything in eithr the man page for mount_nfs *or* nfsd that might account / corect for something like this, but since I'm not sure what this is exactly, not sure exactl what I should be looking for :( Note that this behaviour happens at the *physical* server level as well, having tested with using postalias to generate the same 'lock' issue above ... Now, I do have mountd/nfsd started iwth the -h to bind them to 192.168.1.8 ... *but*, the servers themselves, although on same switch do have different default gateways ... I'm not seeing anything within the man page for, say, rpc.statd/rpc.lockd that allows me to bind it to the 192.168.1.0/24 IP, so is it binding to my public IP instead of my private? So nfsd / mount_nfs can talk find, as they go thorugh 192.168.1.0/24 as desired, but rpc.statd/rpc.lockd are the public IPs and not able to talk to each other? Thx ... ___ freebsd-...@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to freebsd-net-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: nfs error: No route to host when starting apache ...
I just setup an nfs mount between two servers ... ServerA, nfsd on 192.168.1.8 ServerB, nfs client on 192.168.1.7 I have a jail, ServerC, running on 192.168.1.7 ... most operations appear to work, but it looks like 'special files' of a sort aren't working, for when I try and startup Apache, I get: [Fri Apr 01 19:42:02 2011] [emerg] (65)No route to host: couldn't grab the accept mutex When I try and do a 'newaliases', I get: # newaliases postalias: fatal: lock /etc/aliases.db: No route to host Yet, for instance, both MySQL and PostgreSQL are running without any issues ... So, the mount is there, it is readable, it is working ... I can ssh into the jail, I can create files, etc ... I do have rpc.lockd and rpc.statd running on both client / server sides ... Since rpc.lockd and rpc.statd expect to be able to do IP broadcast (same goes for rpcbind), I suspect that might be a problem w.r.t. jails, although I know nothing about how jails work? Oh, and you can use the nolock mount option to avoid use of rpc.lockd and rpc.statd. I'm not seeing anything in eithr the man page for mount_nfs *or* nfsd that might account / corect for something like this, but since I'm not sure what this is exactly, not sure exactl what I should be looking for :( Note that this behaviour happens at the *physical* server level as well, having tested with using postalias to generate the same 'lock' issue above ... Now, I do have mountd/nfsd started iwth the -h to bind them to 192.168.1.8 ... *but*, the servers themselves, although on same switch do have different default gateways ... I'm not seeing anything within the man page for, say, rpc.statd/rpc.lockd that allows me to bind it to the 192.168.1.0/24 IP, so is it binding to my public IP instead of my private? So nfsd / mount_nfs can talk find, as they go thorugh 192.168.1.0/24 as desired, but rpc.statd/rpc.lockd are the public IPs and not able to talk to each other? Thx ... ___ freebsd-...@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to freebsd-net-unsubscr...@freebsd.org ___ freebsd-...@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to freebsd-net-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS setup
n dhert ndhert...@gmail.com wrote: I try to set up NFS between two freebsd-8.1 systems. ... # ps -jaxw | grep rpcbind root 747 1 747 7470 Ss??0:00.01 /usr/sbin/rpcbind ... client# mount server.subdomain.topdomain:/home /mnt (or client# mount XXX.YYY.ZZZ.SSS:/home /mnt) hangs, and after trying for about a minute, responds [tcp] server.subdomain.topdomain:/home: RPCPROG_NFS: RPC: Port mapper failure - RPC: Timed out what is missing to make this work? Any chance a firewall is blocking RPC? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS setup
On 18 November 2010 09:03, per...@pluto.rain.com wrote: n dhert ndhert...@gmail.com wrote: I try to set up NFS between two freebsd-8.1 systems. ... # ps -jaxw | grep rpcbind root 747 1 747 7470 Ss??0:00.01 /usr/sbin/rpcbind ... client# mount server.subdomain.topdomain:/home /mnt (or client# mount XXX.YYY.ZZZ.SSS:/home /mnt) hangs, and after trying for about a minute, responds [tcp] server.subdomain.topdomain:/home: RPCPROG_NFS: RPC: Port mapper failure - RPC: Timed out what is missing to make this work? Any chance a firewall is blocking RPC? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org checkout hosts.allow/deny as well ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Issue
On Mon, 6 Sep 2010, Bill Tillman wrote: I have two LAN segments with a FreeBSD server on each. Server A is 10.0.0.254 Server B is 192.168.0.102 I setup server A has two drives and I setup a share on drive #2 to be shared via NFS with the both networks. I also made a symlink on drive #2 to a folder on drive #1 On server B I can nfs_mount the share on server A and see the symlink. But when I try to access the files in the symlink it shows the link is broken, In other words no files show up. On server A I can see the files in the symlink folder just fine. This is expected NFS behaviour: NFS exports filesystems starting at a given (exported) mount point. While there are many reasons for this, think about the security issues if a user on B could create a symlink on your exported volume (because the origin of the symlink will make no difference to the server) to access any file anywhere on A. If you want both disks 1 and 2 visible, the standard solution is to export and mount both disks on B. If the paths (absolute is easiest, but relative can be made to work) are consistent between A and the mounted image of A's filesystems on B, then your symlinks will work -- that is, if you have this kind of /etc/fstab entry, mounting /disk1 on A to /disk1 on B: A:/disk1/somedir/disk1/somedir A:/disk2/disk2 then a symlink in /disk1/somedir/link pointing to /disk1/something will work just fine. A. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Issue
On Tue, 7 Sep 2010, A. Wright wrote: your symlinks will work -- that is, if you have this kind of /etc/fstab entry, mounting /disk1 on A to /disk1 on B: A:/disk1/somedir/disk1/somedir A:/disk2/disk2 then a symlink in /disk1/somedir/link pointing to /disk1/something will work just fine. That should have read: then a symlink in /disk1/somedir/link pointing to /disk2/something will work just fine A. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: nfs server /home not responding
In response to Lucas Wang lw...@us.toyota-itc.com: We use NFS to store /home directory for users in our lab. However, we occasionally get blocked from logging in because the automount daemon on a NFS client machine hangs. When that happens, we get this error message on the NFS client machine called bucks in its system logs: Aug 24 10:53:40 bucks kernel: nfs server pid...@bucks:/home: not responding pid670 is the amd process. Our NFS server(raptors) has the following configuration: FreeBSD raptors.cs.ucla.edu 7.3-PRERELEASE FreeBSD 7.3-PRERELEASE #0: Tue Feb 9 12:59:50 PST 2010 r...@raptors.cs.ucla.edu:/usr/obj/usr/src/sys/RAPTORS amd64 And the client machine is configured as: FreeBSD bucks.cs.ucla.edu 7.3-PRERELEASE FreeBSD 7.3-PRERELEASE #0: Tue Feb 9 20:47:50 UTC 2010 r...@bucks.cs.ucla.edu:/usr/obj/usr/src/sys/BUCKS amd64 Another thing I want to add is that several other NFS client machines also hang from time to time. But they don't usually hang at the same time. Even though rebooting can fix the problem once, we don't want it keep hurting us. So any insights or suggestions will be greatly appreciated. Thanks a lot. Do you have dumbtimer in the options for the nfs mount? My research into this indicated that the NFS client keeps track of average response times from the server. If the server starts to respond significantly slower than is expected, the code assumes that the server is down and the mount freezes and that message appears in the logs. Usually, after a short wait (a few minutes) the connection resumes and you see a server is alive again message. See man mount_nfs for more info. Also, try switching to TCP mounts. If you have a network that occasionally gets hit with traffic spikes that cause data packets to take abnormally long to travel, or an NFS server that occasionally gets usage spikes that cause it to respond slowly, this will happen. In addition to dumbtimer you can also look at better segmenting your network, or increasing the capacity of the NFS server to prevent the problem. If the NFS hangs occur and the mount never recovers (even after several minutes) then you probably have a different problem. Possibly a firewall is losing the state table and thus the connection is going bad? -- Bill Moran http://www.potentialtech.com http://people.collaborativefusion.com/~wmoran/ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Emergency - RPC and NFSD online but no connection??
Actually a bit more scanning shows Darkice is using 100% of the CPU for some reason? I've commented it out in /etc/rc.conf and initiated a restart so I hope that, the lower load average will make the system start accepting connections again which is a plausible cause for my issue. After testing I will have more information to share, hopefully it will work. If that's the case I will look at limiting CPU horsepower per application so that my lowly system has some room left to compute other things too. On 05/24/2010 12:33 AM, Kaya Saman wrote: Hi guys, this is a really interesting yet annoying issue I'm having. I had a fully working NFS setup until earlier today when I had a failed attempt at recovering an external hard disk that I fitted into my system internally. Now this disk has nothing to do with the system and is used with Linux hence it runs the ext3 filesystem but since the BSD machine is the only desktop I have I thought it a good idea to recover from there. Anyhow, I removed the drive and in the mean time compiled Transmission bit torrent client from ports and a failed attempt to compile Amule2 with a dependency failing to compile; if memory serves me well I think it was cryptopp or something with pp at the end anyway. Sorry for lack of stating on here if it necessary I will find this out and post it immediately. Well just to say now that I can't mount what I used to be able to mount before. I built a little shell script so that I didn't need to use fstab from my Linux box and all I get as response is this: :~# ./BSD2.sh mount.nfs: mount system call failed Syntax in shell script is: mount -t nfs -o rw 172.16.0.200:/mnt/SATA /mnt/BSD2 I have checked the logs on the server /var/log/messages only there isn't any information at all being given?? From BSD if I try to restart or stop nfsd or mountd I get this: rd1# /etc/rc.d/nfsd stop Stopping nfsd. ^C rd1# /etc/rc.d/mountd restart which has held there for a while now meaning that it's probably crashed or something? Netstat claims everything is online: rd1# netstat -ap udp Active Internet connections (including servers) Proto Recv-Q Send-Q Local Address Foreign Address (state) udp4 0 0 *.**.* udp4 0 0 *.tftp *.* udp4 0 0 localhost.ntp *.* udp6 0 0 localhost.ntp *.* udp6 0 0 fe80:3::1.ntp *.* udp4 0 0 rd1.ntp*.* udp6 0 0 *.ntp *.* udp4 0 0 *.ntp *.* udp6 0 0 *.nfsd *.* udp4 0 0 *.nfsd *.* udp4 0 0 *.836 *.* udp6 0 0 *.836 *.* udp6 0 0 *.**.* udp4 0 0 *.653 *.* udp4 0 0 *.sunrpc *.* udp6 0 0 *.760 *.* udp6 0 0 *.sunrpc *.* udp4 0 0 localhost.domain *.* udp4 0 0 rd1.domain *.* udp4 0 0 *.syslog *.* udp6 0 0 *.syslog *.* and I don't have any firewall in place at all!! Output of uname -a: rd1# uname -a FreeBSD rd1.optiplex-networks.com 8.0-RELEASE FreeBSD 8.0-RELEASE #0: Sat Nov 21 15:48:17 UTC 2009 r...@almeida.cse.buffalo.edu:/usr/obj/usr/src/sys/GENERIC i386 The system is a 32bit PIV running at 2.4GHz with 480MB of RAM. Really I'm not sure what to do if I need to upgrade NFS as one of it's dependencies has been upgraded or something else as it's just not working?? Actually I've just typed in exit after su - 'ing to root from an SSH session and the session looks like it's hung on me...?? Also I've had the system running into kernel panic and restarting a lot earlier as the load average went up is what logwatch seems to show. Can anyone help me out of this dilemma?? Regards, Kaya ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Emergency - RPC and NFSD online but no connection??
Ok :-) All is well... That was fast and quick thinking by myself I do say :-P Now to limit Darkice's load on the system?? On 05/24/2010 12:54 AM, Kaya Saman wrote: Actually a bit more scanning shows Darkice is using 100% of the CPU for some reason? I've commented it out in /etc/rc.conf and initiated a restart so I hope that, the lower load average will make the system start accepting connections again which is a plausible cause for my issue. After testing I will have more information to share, hopefully it will work. If that's the case I will look at limiting CPU horsepower per application so that my lowly system has some room left to compute other things too. On 05/24/2010 12:33 AM, Kaya Saman wrote: Hi guys, this is a really interesting yet annoying issue I'm having. I had a fully working NFS setup until earlier today when I had a failed attempt at recovering an external hard disk that I fitted into my system internally. Now this disk has nothing to do with the system and is used with Linux hence it runs the ext3 filesystem but since the BSD machine is the only desktop I have I thought it a good idea to recover from there. Anyhow, I removed the drive and in the mean time compiled Transmission bit torrent client from ports and a failed attempt to compile Amule2 with a dependency failing to compile; if memory serves me well I think it was cryptopp or something with pp at the end anyway. Sorry for lack of stating on here if it necessary I will find this out and post it immediately. Well just to say now that I can't mount what I used to be able to mount before. I built a little shell script so that I didn't need to use fstab from my Linux box and all I get as response is this: :~# ./BSD2.sh mount.nfs: mount system call failed Syntax in shell script is: mount -t nfs -o rw 172.16.0.200:/mnt/SATA /mnt/BSD2 I have checked the logs on the server /var/log/messages only there isn't any information at all being given?? From BSD if I try to restart or stop nfsd or mountd I get this: rd1# /etc/rc.d/nfsd stop Stopping nfsd. ^C rd1# /etc/rc.d/mountd restart which has held there for a while now meaning that it's probably crashed or something? Netstat claims everything is online: rd1# netstat -ap udp Active Internet connections (including servers) Proto Recv-Q Send-Q Local Address Foreign Address (state) udp4 0 0 *.**.* udp4 0 0 *.tftp *.* udp4 0 0 localhost.ntp *.* udp6 0 0 localhost.ntp *.* udp6 0 0 fe80:3::1.ntp *.* udp4 0 0 rd1.ntp*.* udp6 0 0 *.ntp *.* udp4 0 0 *.ntp *.* udp6 0 0 *.nfsd *.* udp4 0 0 *.nfsd *.* udp4 0 0 *.836 *.* udp6 0 0 *.836 *.* udp6 0 0 *.**.* udp4 0 0 *.653 *.* udp4 0 0 *.sunrpc *.* udp6 0 0 *.760 *.* udp6 0 0 *.sunrpc *.* udp4 0 0 localhost.domain *.* udp4 0 0 rd1.domain *.* udp4 0 0 *.syslog *.* udp6 0 0 *.syslog *.* and I don't have any firewall in place at all!! Output of uname -a: rd1# uname -a FreeBSD rd1.optiplex-networks.com 8.0-RELEASE FreeBSD 8.0-RELEASE #0: Sat Nov 21 15:48:17 UTC 2009 r...@almeida.cse.buffalo.edu:/usr/obj/usr/src/sys/GENERIC i386 The system is a 32bit PIV running at 2.4GHz with 480MB of RAM. Really I'm not sure what to do if I need to upgrade NFS as one of it's dependencies has been upgraded or something else as it's just not working?? Actually I've just typed in exit after su - 'ing to root from an SSH session and the session looks like it's hung on me...?? Also I've had the system running into kernel panic and restarting a lot earlier as the load average went up is what logwatch seems to show. Can anyone help me out of this dilemma?? Regards, Kaya ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Mount FreeBSD 8.0
On 04/15/10 15:35, Grant Peel wrote: Hi all, I have been running a backups storage server for many years on FreeBSD 5.2.1. It has been and still is working fine. Several 6.x machines are connected to it on the local network. Since installing FreeBSD 8.0 on two machines (they used to run 6.x and connected to the nfs mount fine), But with FreeBSD 8, the are no longer connecting. ps ax shows (on the client machiens) 551 ?? Is 0:00.00 mount_nfs -t 10 -b -o rw enterprise:/mnt /mnt Just for an experiment, what does showmount -e nfs_server say on the client and the server sides? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Mount FreeBSD 8.0
Ivan, I actually just got it to work. Not sure why the default TCP no longer works but I added the -U flag to the fstab for the mount and it works. Anyone know what may bave changed in FreeBSD 8 to cause this? -Grant P.S on the server machine the output you were looking for was /mnt 192.168.0.0 - Original Message - From: Ivan Voras ivo...@freebsd.org To: freebsd-questions@freebsd.org Sent: Thursday, April 15, 2010 10:19 AM Subject: Re: NFS Mount FreeBSD 8.0 On 04/15/10 15:35, Grant Peel wrote: Hi all, I have been running a backups storage server for many years on FreeBSD 5.2.1. It has been and still is working fine. Several 6.x machines are connected to it on the local network. Since installing FreeBSD 8.0 on two machines (they used to run 6.x and connected to the nfs mount fine), But with FreeBSD 8, the are no longer connecting. ps ax shows (on the client machiens) 551 ?? Is 0:00.00 mount_nfs -t 10 -b -o rw enterprise:/mnt /mnt Just for an experiment, what does showmount -e nfs_server say on the client and the server sides? ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Mount FreeBSD 8.0
On Thu, Apr 15, 2010 at 9:47 AM, Grant Peel gp...@thenetnow.com wrote: Ivan, I actually just got it to work. Not sure why the default TCP no longer works but I added the -U flag to the fstab for the mount and it works. Anyone know what may bave changed in FreeBSD 8 to cause this? -Grant P.S on the server machine the output you were looking for was /mnt 192.168.0.0 Please don't top post. FBSD 8 has a new NFS implementation which might be the cause of your issues. In particular this seems relevant. http://lists.freebsd.org/pipermail/freebsd-current/2009-November/013172.html There's been more than one nfs issue on 8 however so it could easily be something else. 8-STABLE has received a lot of NFS love so you could try that on your clients perhaps. -- Adam Vande More ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
RE: NFS exporting mounted msdosfs subdir
Hello, Found another supporting argument that an MSDOSFS path should be able to be exported through NFS is that -- beside UFS -- CDFS is also working fine. Whom would be the right forum / person to address the below error to? Checked the man for nfsd and no contact is mentioned there. 22:47:45.184593 IP 10.0.0.4.1973235244 10.0.0.2.nfs: 116 readdir [|nfs] 22:47:45.184707 IP 10.0.0.2.nfs 10.0.0.4.1973235244: reply ok 608 readdir 22:47:45.186389 IP 10.0.0.4.1973235245 10.0.0.2.nfs: 120 readdirplus [|nfs] 22:47:45.186499 IP 10.0.0.2.nfs 10.0.0.4.1973235245: reply ok 116 readdirplus ERROR: Operation not supported Thanks, Balazs From: sbre...@hotmail.com CC: freebsd-questions@freebsd.org Date: Thu, 14 Jan 2010 18:46:45 + Subject: RE: NFS exporting mounted msdosfs subdir Hi, A quick search on the internet shows that people use msdosfs with NFS, at least on NetBSD (sorry): http://arkiv.netbsd.se/?ml=dfbsd-bugsa=2004-04t=104901 My FreeBSD mount also shows that the msdosfs mount point is NFS exported. So, from the side of whether nfsd supports msdosfs, I am convinced. Any further idea for this error: 22:47:45.184593 IP 10.0.0.4.1973235244 10.0.0.2.nfs: 116 readdir [|nfs] 22:47:45.184707 IP 10.0.0.2.nfs 10.0.0.4.1973235244: reply ok 608 readdir 22:47:45.186389 IP 10.0.0.4.1973235245 10.0.0.2.nfs: 120 readdirplus [|nfs] 22:47:45.186499 IP 10.0.0.2.nfs 10.0.0.4.1973235245: reply ok 116 readdirplus ERROR: Operation not supported ? -Balazs From: cswi...@mac.com Date: Wed, 13 Jan 2010 14:24:55 -0800 To: sbre...@hotmail.com CC: freebsd-questions@freebsd.org Subject: Re: NFS exporting mounted msdosfs subdir Hi-- On Jan 13, 2010, at 1:52 PM, sbre...@hotmail.com wrote: Anyone has got an idea how this can be resolved? Thanks. Does FreeBSD even support NFS-exporting a locally mounted MS-DOS filesystem? Traditionally, NFS was implemented over the default UFS filesystem and it was common for other filesystem typess to not be exportable -- -Chuck ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org _ Windows Live: Friends get your Flickr, Yelp, and Digg updates when they e-mail you. http://www.microsoft.com/middleeast/windows/windowslive/see-it-in-action/social-network-basics.aspx?ocid=PID23461::T:WLMTAGL:ON:WL:en-xm:SI_SB_3:092010 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org _ Windows Live: Keep your friends up to date with what you do online. http://www.microsoft.com/middleeast/windows/windowslive/see-it-in-action/social-network-basics.aspx?ocid=PID23461::T:WLMTAGL:ON:WL:en-xm:SI_SB_1:092010 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
RE: NFS exporting mounted msdosfs subdir
Hi, A quick search on the internet shows that people use msdosfs with NFS, at least on NetBSD (sorry): http://arkiv.netbsd.se/?ml=dfbsd-bugsa=2004-04t=104901 My FreeBSD mount also shows that the msdosfs mount point is NFS exported. So, from the side of whether nfsd supports msdosfs, I am convinced. Any further idea for this error: 22:47:45.184593 IP 10.0.0.4.1973235244 10.0.0.2.nfs: 116 readdir [|nfs] 22:47:45.184707 IP 10.0.0.2.nfs 10.0.0.4.1973235244: reply ok 608 readdir 22:47:45.186389 IP 10.0.0.4.1973235245 10.0.0.2.nfs: 120 readdirplus [|nfs] 22:47:45.186499 IP 10.0.0.2.nfs 10.0.0.4.1973235245: reply ok 116 readdirplus ERROR: Operation not supported ? -Balazs From: cswi...@mac.com Date: Wed, 13 Jan 2010 14:24:55 -0800 To: sbre...@hotmail.com CC: freebsd-questions@freebsd.org Subject: Re: NFS exporting mounted msdosfs subdir Hi-- On Jan 13, 2010, at 1:52 PM, sbre...@hotmail.com wrote: Anyone has got an idea how this can be resolved? Thanks. Does FreeBSD even support NFS-exporting a locally mounted MS-DOS filesystem? Traditionally, NFS was implemented over the default UFS filesystem and it was common for other filesystem typess to not be exportable -- -Chuck ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org _ Windows Live: Friends get your Flickr, Yelp, and Digg updates when they e-mail you. http://www.microsoft.com/middleeast/windows/windowslive/see-it-in-action/social-network-basics.aspx?ocid=PID23461::T:WLMTAGL:ON:WL:en-xm:SI_SB_3:092010 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
RE: NFS exporting mounted msdosfs subdir
Hello, A mounted msdosfs (USD flash drive) path is exported from FreeBSD. The client (Windows with SFU 3.5) maps the remote NFS path successfully to a local drive (net use ...). However, when the drive is opened (dir ...), the client gets into an endless loop. What I have discovered on FreeBSD is: --- 22:47:45.183215 IP 10.0.0.2.nfs 10.0.0.4.1973235243: reply ok 116 readdirplus ERROR: Operation not supported 22:47:45.184593 IP 10.0.0.4.1973235244 10.0.0.2.nfs: 116 readdir [|nfs] 22:47:45.184707 IP 10.0.0.2.nfs 10.0.0.4.1973235244: reply ok 608 readdir 22:47:45.186389 IP 10.0.0.4.1973235245 10.0.0.2.nfs: 120 readdirplus [|nfs] 22:47:45.186499 IP 10.0.0.2.nfs 10.0.0.4.1973235245: reply ok 116 readdirplus ERROR: Operation not supported 22:47:45.187898 IP 10.0.0.4.1973235246 10.0.0.2.nfs: 116 readdir [|nfs] 22:47:45.188011 IP 10.0.0.2.nfs 10.0.0.4.1973235246: reply ok 608 readdir 22:47:45.189828 IP 10.0.0.4.1973235247 10.0.0.2.nfs: 120 readdirplus [|nfs] 22:47:45.189933 IP 10.0.0.2.nfs 10.0.0.4.1973235247: reply ok 116 readdirplus ERROR: Operation not supported 22:47:45.191358 IP 10.0.0.4.1973235248 10.0.0.2.nfs: 116 readdir [|nfs] --- With no end... Anyone has got an idea how this can be resolved? Thanks. Cheers, Balazs Date: Sun, 10 Jan 2010 04:50:18 -0700 From: fb...@peterk.org To: sbre...@hotmail.com CC: freebsd-questions@freebsd.org Subject: Re: NFS exporting mounted msdosfs subdir Hello, There is an issue with my exported home folder. It has a subdirectory under which an msdosfs pen drive is mounted. The home folder gets exported nicely however the msdos subdirectory is not! Any idea how this could be solved? Some useful extracts: --- babapc# more /etc/fstab # DeviceMountpoint FStype Options DumpPass# /dev/ad4s1b noneswap sw 0 0 /dev/ad4s1a / ufs rw 1 1 /dev/ad4s1e /tmpufs rw 2 2 /dev/ad4s1f /usrufs rw 2 2 /dev/ad4s1d /varufs rw 2 2 /dev/acd0 /cdrom cd9660 ro,noauto 0 0 /dev/da0s1 /usr/home/sbremal/usb_flash_drive msdosfs rw,noauto 0 0 babapc# more /etc/exports #The following examples export /usr to 3 machines named after ducks, #/usr/src and /usr/obj read-only to machines named after trouble makers, #/home and all directories under it to machines named after dead rock stars #and, /a to a network of privileged machines allowed to write on it as root. #/usr huey louie dewie #/usr/src /usr/obj -ro calvin hobbes #/home -alldirs janice jimmy frank #/a -maproot=0 -network 10.0.1.0 -mask 255.255.248.0 # # You should replace these lines with your actual exported filesystems. # Note that BSD's export syntax is 'host-centric' vs. Sun's 'FS-centric' one. /usr/home/sbremal babapc# mount /dev/ad4s1a on / (ufs, local) devfs on /dev (devfs, local) /dev/ad4s1e on /tmp (ufs, local, soft-updates) /dev/ad4s1f on /usr (ufs, NFS exported, local, soft-updates) /dev/ad4s1d on /var (ufs, local, soft-updates) /dev/da0s1 on /usr/home/sbremal/usb_flash_drive (msdosfs, local) 10.0.0.2:/usr/home/sbremal on /root/x (nfs) babapc# ls /root/x/usb_flash_drive/ - Nothing!!! babapc# ls /usr/home/sbremal/usb_flash_drive/ @Nokia Images Backup --- Any help would be much appreciated. (Would freebsd-fs be more appropriate to ask the question on?) Cheers, Balazs You will need to export '/usr/home/sbremal/usb_flash_drive' also since it is another filesystem, not just a subdirectory. [reason you can see the subdirectory, but not the actual FS mounted in there] ]Peter[ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org _ Windows Live: Friends get your Flickr, Yelp, and Digg updates when they e-mail you. http://www.microsoft.com/middleeast/windows/windowslive/see-it-in-action/social-network-basics.aspx?ocid=PID23461::T:WLMTAGL:ON:WL:en-xm:SI_SB_3:092010 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS exporting mounted msdosfs subdir
Hi-- On Jan 13, 2010, at 1:52 PM, sbre...@hotmail.com wrote: Anyone has got an idea how this can be resolved? Thanks. Does FreeBSD even support NFS-exporting a locally mounted MS-DOS filesystem? Traditionally, NFS was implemented over the default UFS filesystem and it was common for other filesystem typess to not be exportable -- -Chuck ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS exporting mounted msdosfs subdir
Hello, There is an issue with my exported home folder. It has a subdirectory under which an msdosfs pen drive is mounted. The home folder gets exported nicely however the msdos subdirectory is not! Any idea how this could be solved? Some useful extracts: --- babapc# more /etc/fstab # Device   Mountpoint FStype Options Dump   Pass# /dev/ad4s1b none   swap   sw 0  0 /dev/ad4s1a /  ufs rw 1  1 /dev/ad4s1e /tmp   ufs rw 2  2 /dev/ad4s1f /usr   ufs rw 2  2 /dev/ad4s1d /var   ufs rw 2  2 /dev/acd0  /cdrom cd9660 ro,noauto  0  0 /dev/da0s1 /usr/home/sbremal/usb_flash_drive  msdosfs rw,noauto  0  0 babapc# more /etc/exports #The following examples export /usr to 3 machines named after ducks, #/usr/src and /usr/obj read-only to machines named after trouble makers, #/home and all directories under it to machines named after dead rock stars #and, /a to a network of privileged machines allowed to write on it as root. #/usr  huey louie dewie #/usr/src /usr/obj -ro calvin hobbes #/home  -alldirs  janice jimmy frank #/a -maproot=0 -network 10.0.1.0 -mask 255.255.248.0 # # You should replace these lines with your actual exported filesystems. # Note that BSD's export syntax is 'host-centric' vs. Sun's 'FS-centric' one. /usr/home/sbremal babapc# mount /dev/ad4s1a on / (ufs, local) devfs on /dev (devfs, local) /dev/ad4s1e on /tmp (ufs, local, soft-updates) /dev/ad4s1f on /usr (ufs, NFS exported, local, soft-updates) /dev/ad4s1d on /var (ufs, local, soft-updates) /dev/da0s1 on /usr/home/sbremal/usb_flash_drive (msdosfs, local) 10.0.0.2:/usr/home/sbremal on /root/x (nfs) babapc# ls /root/x/usb_flash_drive/ - Nothing!!! babapc# ls /usr/home/sbremal/usb_flash_drive/ @Nokia Images Backup --- Any help would be much appreciated. (Would freebsd-fs be more appropriate to ask the question on?) Cheers, Balazs You will need to export '/usr/home/sbremal/usb_flash_drive' also since it is another filesystem, not just a subdirectory. [reason you can see the subdirectory, but not the actual FS mounted in there] ]Peter[ ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS- SAN - FreeBSD
Hi Grant, I'm in a similar situation to where you were in July, and I was wondering what route you ended up going? Patrick On Tue, Jul 21, 2009 at 4:42 PM, Grant Peel gp...@thenetnow.com wrote: Chris, Again, thanks for the info. I only have one server with a PERC (raid) card installed, and I beleive it is an older PERC 3 DCI, and doubt it would do the job. I would not be able to add more PERC cards to the other machines. I am looking to have the connections all done via Ethernet. Again, the connections would be local (device to my switch, switch to the individual servers). Does this mean I should be considering iSCSI, or, since the connections will all be on a local network, that I can continue to consider NFS? Any takers? -Grant - Original Message - From: Christopher J. Umina chris.um...@studsvikscandpower.com To: Grant Peel gp...@thenetnow.com Cc: questi...@freebsd.org Sent: Tuesday, July 21, 2009 8:01 PM Subject: Re: NFS- SAN - FreeBSD Grant, DAS = Direct-Attached Storage, sorry to be confusing. I cannot personally speak to the performance of FreeBSD's NFS, but I wouldn't expect it to be the bottleneck in the situation described. Â Maybe others with more experience could chime in on this topic. The way to use a DAS is to connect the DAS to a server with an external SAS cable (or two). Â The PERC6/E controller you would need inside the server is very well supported in FreeBSD. Â The DAS system would basically act the same as internal disks would act (in the case of the MD1000). Â Of course you'll want to check with Dell before you make any purchases to be positive that your hardware will all communicate nicely, as I'm no Dell salesperson. Depending on how large of an array you plan to make (if larger than 2TB) you may have to investigate gpart/gpt to partition correctly, but that's quite simple in my experience. Chris Grant Peel wrote: Chris, Thanks for the insight! I will defineately investigate that DAS ... although I am not (yet) sure what the acronym means, I am sure it is something akin to Direct Access SCSI. You are quite right, I would like to use NFS to connect the device to the 6 servers I have, again, it would be only hosting the /home partition for each of them. Do you know if there would be any NFS I/O slowdowns using it in that fassion? Would freebsd support (on the storage device) that many connections? Also, do the Dell DAS machines run with FreeBSD? Also, from you you explained, I doubt I really need the versatility of the SAN at this point, or in the near future. I simply want a mass /home storage unit. -Grant - Original Message - From: Christopher J. Umina chris.um...@studsvikscandpower.com To: Grant Peel gp...@thenetnow.com Cc: questi...@freebsd.org Sent: Tuesday, July 21, 2009 5:43 PM Subject: Re: NFS- SAN - FreeBSD Grant, I mean to say that often times external SCSI solutions (direct attached) are cheaper and perform better (in terms of I/O) than iSCSI SANs. Especially if you're using many disks. Â SANs are generally chosen for the ability to be split into LUNs for different servers. Â Think of it as a disk which you can partition and serve out to servers on a per-partition basis, over Ethernet. Â That's essentially what an iSCSI SAN does. Â While DAS systems allow the same sort of configuration, they don't serve out over Ethernet, only SCSI/SAS. Since you plan to use NFS to share the files to the other servers, I think it may make more sense for you to use a SCSI solution if yo don't need the versatility of a SAN. Of course I know nothing of how you plan to expand this system, but from what I understand, with Dell DAS hardware it is possible to connect up to 4 different servers to the DAS and expand to up to 6 15 disk enclosures. The MD3000i (iSCSI) expands only to 3. Another issue is that without compiling in special versions of the iSCSI initiator, even in 8.0-BETA2 (which is not production-ready), iSCSI performance and reliability are terrible. Â There are other versions of the code (which I currently use) for the iscsi_initiator kernel module, but unless you're comfortable doing that, you may consider DAS in terms of ease of implementation and maintenance as well. Chris Grant Peel wrote: Chris, I don't know what a direct attached array is. What I was just thinking was move all of the servers /home directory to a huge NFS mount. If you have the time to elaborate fursther, I would apprciate it... This iSCSI think has me entrigued, but I must admit I know little about it at this point. -Grant - Original Message - From: Christopher J. Umina chris.um...@studsvik.com To: Grant Peel gp...@thenetnow.com Sent: Monday, July 20, 2009 11:27 PM Subject: Re: NFS- SAN - FreeBSD Grant, I have to ask, is there a reason you're intent on going with a SAN versus a direct-attached array? Chris Grant Peel wrote: Thanks
Re: NFS and crossmount
On Monday 16 November 2009 06:10:23 Patrik Usher wrote: I'm chaning fileserver to a FreeBSD 7.2 from my old linux and can't find how to define the option crossmnt (crossmount) for NFS. Does anyone know if it's supported under FreeBSD 7 and if so, how to define it ? I don't believe a similar option is available. You need a line in /etc/exports for each filesystem (mountpoint) you wish to export. JN ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS performance-tuning FreeBSD - NetApp
Merhaba Ewald, You can read http://communities.netapp.com/thread/39 thread. There is special mount options for Linux and also for FreeBSD give it a try. Regards. Monday, August 3, 2009, 11:55:16 AM, you wrote: Hi, I've got a FreeBSD 7.2 box (HP C-class Blade - AMD dual core Opteron (x64), 4GB RAM, Broadcom NetXtreme II BCM5706) that should be connected to a NetApp 3170 filer via NFS. Out of the box, with nothing tuned (no special parameters for mount_nfs, no kernel tuning), performance is very sluggish: I've got ~250Mbit/sec performance with peaks around 400Mbit/sec. Sure enough, neither CPU (server and NetApp) nor network performance is the problem here - it must be something NFS-related. Any ideas on how to increas my NFS-performance? (Special mount parameters, kernel tuning,...) Thanks in advance for any clue, -ewald ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org -- Best regards, Omermailto:of...@enderunix.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS performance-tuning FreeBSD - NetApp
On Aug 3, 2009, at 4:55 AM, Ewald Jenisch wrote: Hi, I've got a FreeBSD 7.2 box (HP C-class Blade - AMD dual core Opteron (x64), 4GB RAM, Broadcom NetXtreme II BCM5706) that should be connected to a NetApp 3170 filer via NFS. Out of the box, with nothing tuned (no special parameters for mount_nfs, no kernel tuning), performance is very sluggish: I've got ~250Mbit/sec performance with peaks around 400Mbit/sec. Sure enough, neither CPU (server and NetApp) nor network performance is the problem here - it must be something NFS-related. Any ideas on how to increas my NFS-performance? (Special mount parameters, kernel tuning,...) I would suggest bumping the read and write sizes to 32K and using tcp instead of udp If you have very large directories, you can also see an increase in responsiveness by enabling readdirplus as well, but that wont help with raw throughput. Try passing the following parameters to mount and see if performance is any better -r=32768 -w=32768 -l -T -- Steven Kreuzer http://www.exit2shell.com/~skreuzer ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
RE: NFS : -alldirs requested but is not a filesystem mountpoint - SOLVED
Solved: -alldirs was unnecessary Thanks. -Mensaje original- De: owner-freebsd-questi...@freebsd.org [mailto:owner-freebsd-questi...@freebsd.org] En nombre de Aitor San Juan Enviado el: lunes, 27 de julio de 2009 9:26 Para: freebsd-questions@freebsd.org Asunto: NFS : -alldirs requested but is not a filesystem mountpoint Hello List, I have a FreeBSD 5.4 (yes a bit old), and I'm just using it as a NFS server. The server has got a filesystem located at /data This local filesystem has several subdirectories, and I'd like these subdirectories to be visible to a client host, but not all as a whole. This is the hierarchy at /data: /data/bulletins/nfs_bulls /data/bulletins/etc/ftpmotd -- this is only for FTP's ftpchroot config /data/taxes/docs /data/taxes/etc/ftpmotd -- this is only for FTP's ftpchroot config The data to be visible is under /data/bulletins/nfs_bulls (and all its subdirectories) and /data/taxes/docs (and all its subdirectories) As the content of both directories has nothing to do with each other, I'd like them to be independently exported. Thus, a client needing access to these data should issue 2 mount's for each data. I don't want both directories to be visible with just one mount. This is what I've coded in my /etc/exports /data/bulletins/nfs_bulls -alldirs -ro -network MyNet -mask 255.255.255.0 /data/taxes/docs -alldirs -ro -network MyNet -mask 255.255.255.0 And this is what syslog says: -alldirs requested but /data/bulletins/nfs_bulls is not a filesystem mountpoint bad exports list line /data/bulletins/nfs_bulls -alldirs -ro -network MyNet -mask 255.255.255.0 -alldirs requested but /data/taxes/docs is not a filesystem mountpoint bad exports list line /data/taxes/docs -alldirs -ro -network MyNet -mask 255.255.255.0 Reading through etc/exports(5) and having seen those error messages, I think I should create another separate filesystem on my server. However, I haven't got any space left for another slice. Is it possible to somehow accomplish what I want in this scenario? Many thanks in advance. LEGEZKO OHARRA / AVISO LEGAL / LEGAL ADVICE * Mezu honek isilpeko informazioa gorde dezake, edo jabea duena, edota legez babestuta dagoena. Zuri zuzendua ez bada, bidali duenari esan eta ezabatu, inori berbidali edo gorde gabe, legeak debekatzen duelako mezuak erabiltzea baimenik gabe. -- Este mensaje puede contener informacion confidencial, en propiedad o legalmente protegida. Si usted no es el destinatario, le rogamos lo comunique al remitente y proceda a borrarlo, sin reenviarlo ni conservarlo, ya que su uso no autorizado esta prohibido legalmente. -- This message may contain confidential, proprietary or legally privileged information. If you are not the intended recipient of this message, please notify it to the sender and delete without resending or backing it, as it is legally prohibited. ** ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS- SAN - FreeBSD
Grant, I mean to say that often times external SCSI solutions (direct attached) are cheaper and perform better (in terms of I/O) than iSCSI SANs. Especially if you're using many disks. SANs are generally chosen for the ability to be split into LUNs for different servers. Think of it as a disk which you can partition and serve out to servers on a per-partition basis, over Ethernet. That's essentially what an iSCSI SAN does. While DAS systems allow the same sort of configuration, they don't serve out over Ethernet, only SCSI/SAS. Since you plan to use NFS to share the files to the other servers, I think it may make more sense for you to use a SCSI solution if yo don't need the versatility of a SAN. Of course I know nothing of how you plan to expand this system, but from what I understand, with Dell DAS hardware it is possible to connect up to 4 different servers to the DAS and expand to up to 6 15 disk enclosures. The MD3000i (iSCSI) expands only to 3. Another issue is that without compiling in special versions of the iSCSI initiator, even in 8.0-BETA2 (which is not production-ready), iSCSI performance and reliability are terrible. There are other versions of the code (which I currently use) for the iscsi_initiator kernel module, but unless you're comfortable doing that, you may consider DAS in terms of ease of implementation and maintenance as well. Chris Grant Peel wrote: Chris, I don't know what a direct attached array is. What I was just thinking was move all of the servers /home directory to a huge NFS mount. If you have the time to elaborate fursther, I would apprciate it... This iSCSI think has me entrigued, but I must admit I know little about it at this point. -Grant - Original Message - From: Christopher J. Umina chris.um...@studsvik.com To: Grant Peel gp...@thenetnow.com Sent: Monday, July 20, 2009 11:27 PM Subject: Re: NFS- SAN - FreeBSD Grant, I have to ask, is there a reason you're intent on going with a SAN versus a direct-attached array? Chris Grant Peel wrote: Thanks for the reply. I have not used/investigated the iSCSI thing yet The original question is can I just use an NFS mount to the storage's /home partition? -Grant - Original Message - From: mojo fms To: Grant Peel Cc: freebsd-questions@freebsd.org Sent: Monday, July 20, 2009 4:21 PM Subject: Re: NFS- SAN - FreeBSD You would be better off at least having the SAN on 1gb ethernet or even better tripple 1gb (on a 100mb switch should be fine but you need failover for higher avaliability) ethernet for latency and failover reasons with a hot backup on the network controller. I dont see why you could not do this, its just iscsi connection normally so there is not a big issue getting freebsd to connect to it. We run 2 of the 16tb powervault which does pretty well for storage, one runs everything and the other is a replicated offsite backup. Performance wise, it really depends on how many servers you have pulling data from the SAN and how hard the IO works on the current servers. If you have 100 servers you might push the IO a bit but but it should be fine if your not serving more than 2Mb/s out to everyone, the servers and disks are going to cache a fair amount of always used data. On Mon, Jul 20, 2009 at 11:52 AM, Grant Peel gp...@thenetnow.com wrote: Hi all, I am assuming by the lack of response, my question to too long winded, let me re-phrase: What kind of performance might I expect if I load FreeBSD 7.2 on a 24 disk, Dell PowerVault when its only mission is to serve as a local area storage unit (/home). Obviously, to store all users /home data. Throug an NFS connection via fast (100m/b) ethernet. Each connecting server (6) contain about 200 domains? -Grant - Original Message - From: Grant Peel gp...@thenetnow.com To: freebsd-questions@freebsd.org Sent: Saturday, July 18, 2009 10:35 AM Subject: NFS- SAN - FreeBSD Hi all, Up to this point, all of our servers are standalone, i.e. all services and software required are installed on each local server. Apache, Exim, vm-pop3d, Mysql, etc etc. Each local server is connected to the Inet via a VLAN (WAN), to our colo's switch. Each server contains about 300 domains, each domain has its own IP. Each sever is also connected to a VLAN (LAN) via the same (Dell 48 Port managed switch). We have been considering consolidating all users data from each server to a central (local), storage unit. While I do have active nfs's running (for backups etc), on the LAN only, I have never attempted to create 1 mass storage unit. So I suppose the questions are: 1) Is there any specific hardware that anyone might reccommend? I want to stick with FreeBSD as the OS as I am quite comfortable admining it, 2) Would anyone reccomend NOT using FreeBSD? Why
Re: NFS- SAN - FreeBSD
Grant, DAS = Direct-Attached Storage, sorry to be confusing. I cannot personally speak to the performance of FreeBSD's NFS, but I wouldn't expect it to be the bottleneck in the situation described. Maybe others with more experience could chime in on this topic. The way to use a DAS is to connect the DAS to a server with an external SAS cable (or two). The PERC6/E controller you would need inside the server is very well supported in FreeBSD. The DAS system would basically act the same as internal disks would act (in the case of the MD1000). Of course you'll want to check with Dell before you make any purchases to be positive that your hardware will all communicate nicely, as I'm no Dell salesperson. Depending on how large of an array you plan to make (if larger than 2TB) you may have to investigate gpart/gpt to partition correctly, but that's quite simple in my experience. Chris Grant Peel wrote: Chris, Thanks for the insight! I will defineately investigate that DAS ... although I am not (yet) sure what the acronym means, I am sure it is something akin to Direct Access SCSI. You are quite right, I would like to use NFS to connect the device to the 6 servers I have, again, it would be only hosting the /home partition for each of them. Do you know if there would be any NFS I/O slowdowns using it in that fassion? Would freebsd support (on the storage device) that many connections? Also, do the Dell DAS machines run with FreeBSD? Also, from you you explained, I doubt I really need the versatility of the SAN at this point, or in the near future. I simply want a mass /home storage unit. -Grant - Original Message - From: Christopher J. Umina chris.um...@studsvikscandpower.com To: Grant Peel gp...@thenetnow.com Cc: questi...@freebsd.org Sent: Tuesday, July 21, 2009 5:43 PM Subject: Re: NFS- SAN - FreeBSD Grant, I mean to say that often times external SCSI solutions (direct attached) are cheaper and perform better (in terms of I/O) than iSCSI SANs. Especially if you're using many disks. SANs are generally chosen for the ability to be split into LUNs for different servers. Think of it as a disk which you can partition and serve out to servers on a per-partition basis, over Ethernet. That's essentially what an iSCSI SAN does. While DAS systems allow the same sort of configuration, they don't serve out over Ethernet, only SCSI/SAS. Since you plan to use NFS to share the files to the other servers, I think it may make more sense for you to use a SCSI solution if yo don't need the versatility of a SAN. Of course I know nothing of how you plan to expand this system, but from what I understand, with Dell DAS hardware it is possible to connect up to 4 different servers to the DAS and expand to up to 6 15 disk enclosures. The MD3000i (iSCSI) expands only to 3. Another issue is that without compiling in special versions of the iSCSI initiator, even in 8.0-BETA2 (which is not production-ready), iSCSI performance and reliability are terrible. There are other versions of the code (which I currently use) for the iscsi_initiator kernel module, but unless you're comfortable doing that, you may consider DAS in terms of ease of implementation and maintenance as well. Chris Grant Peel wrote: Chris, I don't know what a direct attached array is. What I was just thinking was move all of the servers /home directory to a huge NFS mount. If you have the time to elaborate fursther, I would apprciate it... This iSCSI think has me entrigued, but I must admit I know little about it at this point. -Grant - Original Message - From: Christopher J. Umina chris.um...@studsvik.com To: Grant Peel gp...@thenetnow.com Sent: Monday, July 20, 2009 11:27 PM Subject: Re: NFS- SAN - FreeBSD Grant, I have to ask, is there a reason you're intent on going with a SAN versus a direct-attached array? Chris Grant Peel wrote: Thanks for the reply. I have not used/investigated the iSCSI thing yet The original question is can I just use an NFS mount to the storage's /home partition? -Grant - Original Message - From: mojo fms To: Grant Peel Cc: freebsd-questions@freebsd.org Sent: Monday, July 20, 2009 4:21 PM Subject: Re: NFS- SAN - FreeBSD You would be better off at least having the SAN on 1gb ethernet or even better tripple 1gb (on a 100mb switch should be fine but you need failover for higher avaliability) ethernet for latency and failover reasons with a hot backup on the network controller. I dont see why you could not do this, its just iscsi connection normally so there is not a big issue getting freebsd to connect to it. We run 2 of the 16tb powervault which does pretty well for storage, one runs everything and the other is a replicated offsite backup. Performance wise, it really depends on how many servers you have pulling data from the SAN and how hard the IO works on the current servers. If you
Re: NFS- SAN - FreeBSD
Chris, Thanks for the insight! I will defineately investigate that DAS ... although I am not (yet) sure what the acronym means, I am sure it is something akin to Direct Access SCSI. You are quite right, I would like to use NFS to connect the device to the 6 servers I have, again, it would be only hosting the /home partition for each of them. Do you know if there would be any NFS I/O slowdowns using it in that fassion? Would freebsd support (on the storage device) that many connections? Also, do the Dell DAS machines run with FreeBSD? Also, from you you explained, I doubt I really need the versatility of the SAN at this point, or in the near future. I simply want a mass /home storage unit. -Grant - Original Message - From: Christopher J. Umina chris.um...@studsvikscandpower.com To: Grant Peel gp...@thenetnow.com Cc: questi...@freebsd.org Sent: Tuesday, July 21, 2009 5:43 PM Subject: Re: NFS- SAN - FreeBSD Grant, I mean to say that often times external SCSI solutions (direct attached) are cheaper and perform better (in terms of I/O) than iSCSI SANs. Especially if you're using many disks. SANs are generally chosen for the ability to be split into LUNs for different servers. Think of it as a disk which you can partition and serve out to servers on a per-partition basis, over Ethernet. That's essentially what an iSCSI SAN does. While DAS systems allow the same sort of configuration, they don't serve out over Ethernet, only SCSI/SAS. Since you plan to use NFS to share the files to the other servers, I think it may make more sense for you to use a SCSI solution if yo don't need the versatility of a SAN. Of course I know nothing of how you plan to expand this system, but from what I understand, with Dell DAS hardware it is possible to connect up to 4 different servers to the DAS and expand to up to 6 15 disk enclosures. The MD3000i (iSCSI) expands only to 3. Another issue is that without compiling in special versions of the iSCSI initiator, even in 8.0-BETA2 (which is not production-ready), iSCSI performance and reliability are terrible. There are other versions of the code (which I currently use) for the iscsi_initiator kernel module, but unless you're comfortable doing that, you may consider DAS in terms of ease of implementation and maintenance as well. Chris Grant Peel wrote: Chris, I don't know what a direct attached array is. What I was just thinking was move all of the servers /home directory to a huge NFS mount. If you have the time to elaborate fursther, I would apprciate it... This iSCSI think has me entrigued, but I must admit I know little about it at this point. -Grant - Original Message - From: Christopher J. Umina chris.um...@studsvik.com To: Grant Peel gp...@thenetnow.com Sent: Monday, July 20, 2009 11:27 PM Subject: Re: NFS- SAN - FreeBSD Grant, I have to ask, is there a reason you're intent on going with a SAN versus a direct-attached array? Chris Grant Peel wrote: Thanks for the reply. I have not used/investigated the iSCSI thing yet The original question is can I just use an NFS mount to the storage's /home partition? -Grant - Original Message - From: mojo fms To: Grant Peel Cc: freebsd-questions@freebsd.org Sent: Monday, July 20, 2009 4:21 PM Subject: Re: NFS- SAN - FreeBSD You would be better off at least having the SAN on 1gb ethernet or even better tripple 1gb (on a 100mb switch should be fine but you need failover for higher avaliability) ethernet for latency and failover reasons with a hot backup on the network controller. I dont see why you could not do this, its just iscsi connection normally so there is not a big issue getting freebsd to connect to it. We run 2 of the 16tb powervault which does pretty well for storage, one runs everything and the other is a replicated offsite backup. Performance wise, it really depends on how many servers you have pulling data from the SAN and how hard the IO works on the current servers. If you have 100 servers you might push the IO a bit but but it should be fine if your not serving more than 2Mb/s out to everyone, the servers and disks are going to cache a fair amount of always used data. On Mon, Jul 20, 2009 at 11:52 AM, Grant Peel gp...@thenetnow.com wrote: Hi all, I am assuming by the lack of response, my question to too long winded, let me re-phrase: What kind of performance might I expect if I load FreeBSD 7.2 on a 24 disk, Dell PowerVault when its only mission is to serve as a local area storage unit (/home). Obviously, to store all users /home data. Throug an NFS connection via fast (100m/b) ethernet. Each connecting server (6) contain about 200 domains? -Grant - Original Message - From: Grant Peel gp...@thenetnow.com To: freebsd-questions@freebsd.org Sent: Saturday, July 18, 2009 10:35 AM Subject: NFS- SAN - FreeBSD Hi all
Re: NFS- SAN - FreeBSD
Chris, Again, thanks for the info. I only have one server with a PERC (raid) card installed, and I beleive it is an older PERC 3 DCI, and doubt it would do the job. I would not be able to add more PERC cards to the other machines. I am looking to have the connections all done via Ethernet. Again, the connections would be local (device to my switch, switch to the individual servers). Does this mean I should be considering iSCSI, or, since the connections will all be on a local network, that I can continue to consider NFS? Any takers? -Grant - Original Message - From: Christopher J. Umina chris.um...@studsvikscandpower.com To: Grant Peel gp...@thenetnow.com Cc: questi...@freebsd.org Sent: Tuesday, July 21, 2009 8:01 PM Subject: Re: NFS- SAN - FreeBSD Grant, DAS = Direct-Attached Storage, sorry to be confusing. I cannot personally speak to the performance of FreeBSD's NFS, but I wouldn't expect it to be the bottleneck in the situation described. Maybe others with more experience could chime in on this topic. The way to use a DAS is to connect the DAS to a server with an external SAS cable (or two). The PERC6/E controller you would need inside the server is very well supported in FreeBSD. The DAS system would basically act the same as internal disks would act (in the case of the MD1000). Of course you'll want to check with Dell before you make any purchases to be positive that your hardware will all communicate nicely, as I'm no Dell salesperson. Depending on how large of an array you plan to make (if larger than 2TB) you may have to investigate gpart/gpt to partition correctly, but that's quite simple in my experience. Chris Grant Peel wrote: Chris, Thanks for the insight! I will defineately investigate that DAS ... although I am not (yet) sure what the acronym means, I am sure it is something akin to Direct Access SCSI. You are quite right, I would like to use NFS to connect the device to the 6 servers I have, again, it would be only hosting the /home partition for each of them. Do you know if there would be any NFS I/O slowdowns using it in that fassion? Would freebsd support (on the storage device) that many connections? Also, do the Dell DAS machines run with FreeBSD? Also, from you you explained, I doubt I really need the versatility of the SAN at this point, or in the near future. I simply want a mass /home storage unit. -Grant - Original Message - From: Christopher J. Umina chris.um...@studsvikscandpower.com To: Grant Peel gp...@thenetnow.com Cc: questi...@freebsd.org Sent: Tuesday, July 21, 2009 5:43 PM Subject: Re: NFS- SAN - FreeBSD Grant, I mean to say that often times external SCSI solutions (direct attached) are cheaper and perform better (in terms of I/O) than iSCSI SANs. Especially if you're using many disks. SANs are generally chosen for the ability to be split into LUNs for different servers. Think of it as a disk which you can partition and serve out to servers on a per-partition basis, over Ethernet. That's essentially what an iSCSI SAN does. While DAS systems allow the same sort of configuration, they don't serve out over Ethernet, only SCSI/SAS. Since you plan to use NFS to share the files to the other servers, I think it may make more sense for you to use a SCSI solution if yo don't need the versatility of a SAN. Of course I know nothing of how you plan to expand this system, but from what I understand, with Dell DAS hardware it is possible to connect up to 4 different servers to the DAS and expand to up to 6 15 disk enclosures. The MD3000i (iSCSI) expands only to 3. Another issue is that without compiling in special versions of the iSCSI initiator, even in 8.0-BETA2 (which is not production-ready), iSCSI performance and reliability are terrible. There are other versions of the code (which I currently use) for the iscsi_initiator kernel module, but unless you're comfortable doing that, you may consider DAS in terms of ease of implementation and maintenance as well. Chris Grant Peel wrote: Chris, I don't know what a direct attached array is. What I was just thinking was move all of the servers /home directory to a huge NFS mount. If you have the time to elaborate fursther, I would apprciate it... This iSCSI think has me entrigued, but I must admit I know little about it at this point. -Grant - Original Message - From: Christopher J. Umina chris.um...@studsvik.com To: Grant Peel gp...@thenetnow.com Sent: Monday, July 20, 2009 11:27 PM Subject: Re: NFS- SAN - FreeBSD Grant, I have to ask, is there a reason you're intent on going with a SAN versus a direct-attached array? Chris Grant Peel wrote: Thanks for the reply. I have not used/investigated the iSCSI thing yet The original question is can I just use an NFS mount to the storage's /home partition? -Grant - Original Message - From: mojo fms To: Grant Peel Cc: freebsd
Re: NFS- SAN - FreeBSD
Hi all, I am assuming by the lack of response, my question to too long winded, let me re-phrase: What kind of performance might I expect if I load FreeBSD 7.2 on a 24 disk, Dell PowerVault when its only mission is to serve as a local area storage unit (/home). Obviously, to store all users /home data. Throug an NFS connection via fast (100m/b) ethernet. Each connecting server (6) contain about 200 domains? -Grant - Original Message - From: Grant Peel gp...@thenetnow.com To: freebsd-questions@freebsd.org Sent: Saturday, July 18, 2009 10:35 AM Subject: NFS- SAN - FreeBSD Hi all, Up to this point, all of our servers are standalone, i.e. all services and software required are installed on each local server. Apache, Exim, vm-pop3d, Mysql, etc etc. Each local server is connected to the Inet via a VLAN (WAN), to our colo's switch. Each server contains about 300 domains, each domain has its own IP. Each sever is also connected to a VLAN (LAN) via the same (Dell 48 Port managed switch). We have been considering consolidating all users data from each server to a central (local), storage unit. While I do have active nfs's running (for backups etc), on the LAN only, I have never attempted to create 1 mass storage unit. So I suppose the questions are: 1) Is there any specific hardware that anyone might reccommend? I want to stick with FreeBSD as the OS as I am quite comfortable admining it, 2) Would anyone reccomend NOT using FreeBSD? Why? 3) Assuming I am using FreeBSD as the storage systems OS, could NFS simply be used? 4) Considering out whole Inet traffic runs about 2 Mb/s, is there any reason the port to the Storage unit should be more than 100 M/b (would it be imparative to use 1 G/b transfer)? TIA, -Grant ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS- SAN - FreeBSD
Thanks for the reply. I have not used/investigated the iSCSI thing yet The original question is can I just use an NFS mount to the storage's /home partition? -Grant - Original Message - From: mojo fms To: Grant Peel Cc: freebsd-questions@freebsd.org Sent: Monday, July 20, 2009 4:21 PM Subject: Re: NFS- SAN - FreeBSD You would be better off at least having the SAN on 1gb ethernet or even better tripple 1gb (on a 100mb switch should be fine but you need failover for higher avaliability) ethernet for latency and failover reasons with a hot backup on the network controller. I dont see why you could not do this, its just iscsi connection normally so there is not a big issue getting freebsd to connect to it. We run 2 of the 16tb powervault which does pretty well for storage, one runs everything and the other is a replicated offsite backup. Performance wise, it really depends on how many servers you have pulling data from the SAN and how hard the IO works on the current servers. If you have 100 servers you might push the IO a bit but but it should be fine if your not serving more than 2Mb/s out to everyone, the servers and disks are going to cache a fair amount of always used data. On Mon, Jul 20, 2009 at 11:52 AM, Grant Peel gp...@thenetnow.com wrote: Hi all, I am assuming by the lack of response, my question to too long winded, let me re-phrase: What kind of performance might I expect if I load FreeBSD 7.2 on a 24 disk, Dell PowerVault when its only mission is to serve as a local area storage unit (/home). Obviously, to store all users /home data. Throug an NFS connection via fast (100m/b) ethernet. Each connecting server (6) contain about 200 domains? -Grant - Original Message - From: Grant Peel gp...@thenetnow.com To: freebsd-questions@freebsd.org Sent: Saturday, July 18, 2009 10:35 AM Subject: NFS- SAN - FreeBSD Hi all, Up to this point, all of our servers are standalone, i.e. all services and software required are installed on each local server. Apache, Exim, vm-pop3d, Mysql, etc etc. Each local server is connected to the Inet via a VLAN (WAN), to our colo's switch. Each server contains about 300 domains, each domain has its own IP. Each sever is also connected to a VLAN (LAN) via the same (Dell 48 Port managed switch). We have been considering consolidating all users data from each server to a central (local), storage unit. While I do have active nfs's running (for backups etc), on the LAN only, I have never attempted to create 1 mass storage unit. So I suppose the questions are: 1) Is there any specific hardware that anyone might reccommend? I want to stick with FreeBSD as the OS as I am quite comfortable admining it, 2) Would anyone reccomend NOT using FreeBSD? Why? 3) Assuming I am using FreeBSD as the storage systems OS, could NFS simply be used? 4) Considering out whole Inet traffic runs about 2 Mb/s, is there any reason the port to the Storage unit should be more than 100 M/b (would it be imparative to use 1 G/b transfer)? TIA, -Grant ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org -- Who knew ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS- SAN - FreeBSD
You would be better off at least having the SAN on 1gb ethernet or even better tripple 1gb (on a 100mb switch should be fine but you need failover for higher avaliability) ethernet for latency and failover reasons with a hot backup on the network controller. I dont see why you could not do this, its just iscsi connection normally so there is not a big issue getting freebsd to connect to it. We run 2 of the 16tb powervault which does pretty well for storage, one runs everything and the other is a replicated offsite backup. Performance wise, it really depends on how many servers you have pulling data from the SAN and how hard the IO works on the current servers. If you have 100 servers you might push the IO a bit but but it should be fine if your not serving more than 2Mb/s out to everyone, the servers and disks are going to cache a fair amount of always used data. On Mon, Jul 20, 2009 at 11:52 AM, Grant Peel gp...@thenetnow.com wrote: Hi all, I am assuming by the lack of response, my question to too long winded, let me re-phrase: What kind of performance might I expect if I load FreeBSD 7.2 on a 24 disk, Dell PowerVault when its only mission is to serve as a local area storage unit (/home). Obviously, to store all users /home data. Throug an NFS connection via fast (100m/b) ethernet. Each connecting server (6) contain about 200 domains? -Grant - Original Message - From: Grant Peel gp...@thenetnow.com To: freebsd-questions@freebsd.org Sent: Saturday, July 18, 2009 10:35 AM Subject: NFS- SAN - FreeBSD Hi all, Up to this point, all of our servers are standalone, i.e. all services and software required are installed on each local server. Apache, Exim, vm-pop3d, Mysql, etc etc. Each local server is connected to the Inet via a VLAN (WAN), to our colo's switch. Each server contains about 300 domains, each domain has its own IP. Each sever is also connected to a VLAN (LAN) via the same (Dell 48 Port managed switch). We have been considering consolidating all users data from each server to a central (local), storage unit. While I do have active nfs's running (for backups etc), on the LAN only, I have never attempted to create 1 mass storage unit. So I suppose the questions are: 1) Is there any specific hardware that anyone might reccommend? I want to stick with FreeBSD as the OS as I am quite comfortable admining it, 2) Would anyone reccomend NOT using FreeBSD? Why? 3) Assuming I am using FreeBSD as the storage systems OS, could NFS simply be used? 4) Considering out whole Inet traffic runs about 2 Mb/s, is there any reason the port to the Storage unit should be more than 100 M/b (would it be imparative to use 1 G/b transfer)? TIA, -Grant ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org -- Who knew ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
RE: NFS slow
Jan Catrysse wrote: Hello all, I am having some problems with NFS and slow performance. This is the scenario: 2x FreeBSD 7.1. (Raid storage server, MP, the works) GB Lan interface between them. When I transfer 1 big file the speed is never higher than 10MB/s with a peak to 14MB/s. When I transfer multiple files at the same time speed is about 10MB/s per thread. Disk speed 100MB/s Network speed using samba 60MB/s (limited by clients disk speed) Tried enabling NFSlockd, NFSstatd but that changes nothing. Any help or hunch would be greatly appreciated. Here are some ideas for testing: * Any firewall in between them? Do you have network errors? * Any other network problems, like DNS lookup failures? (not that it should matter for sustained tranfers but still...) * Are you using TCP or UDP for NFS? TCP should be better in all cases. * Have you monitored the system with top? Try hitting S and H in top while transfering files, see if anything looks suspicious. * Run iostat 1, check tps and KB/t. * What file system are you using? Hello Ivoras, NFS TCP did the trick. I tried it already but I didn't properly dismount the volumes before remounting them on TCP. I did a mount -u -a instead. Using netstat it became clear NFS was still using UDP. A umount and mount -a did the trick! Thnx! Jan ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS slow
Jan Catrysse wrote: Hello all, I am having some problems with NFS and slow performance. This is the scenario: 2x FreeBSD 7.1. (Raid storage server, MP, the works) GB Lan interface between them. When I transfer 1 big file the speed is never higher than 10MB/s with a peak to 14MB/s. When I transfer multiple files at the same time speed is about 10MB/s per thread. Disk speed 100MB/s Network speed using samba 60MB/s (limited by clients disk speed) Tried enabling NFSlockd, NFSstatd but that changes nothing. Any help or hunch would be greatly appreciated. Here are some ideas for testing: * Any firewall in between them? Do you have network errors? * Any other network problems, like DNS lookup failures? (not that it should matter for sustained tranfers but still...) * Are you using TCP or UDP for NFS? TCP should be better in all cases. * Have you monitored the system with top? Try hitting S and H in top while transfering files, see if anything looks suspicious. * Run iostat 1, check tps and KB/t. * What file system are you using? signature.asc Description: OpenPGP digital signature
Re: NFS, how to find out which files are used
In the last episode (Feb 03), Sandra Kachelmann said: I have an NFS fileserver and would like to figure out which files are being read/written to. Is there something to find that out? Something similar to samba's 'smbstatus' command. The best you can do currently is run tcpdump/wireshark and watch the remote file operations as they happen... NFS doesn't access files by filename, but by NFS filehandle (basically device+inode number), so a remote client first looks up the filename to get the filehandle, and all accesses are done via the filehandle at that point. Theoretically, one could write a dtrace script that watches calls to nfs_namei, nfsrv_read, and nfsrv_write, and then matches read/write ops with the filenames that were looked up beforehand. -- Dan Nelson dnel...@allantgroup.com ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS, how to find out which files are used
2009/2/3 Dan Nelson dnel...@allantgroup.com In the last episode (Feb 03), Sandra Kachelmann said: I have an NFS fileserver and would like to figure out which files are being read/written to. Is there something to find that out? Something similar to samba's 'smbstatus' command. The best you can do currently is run tcpdump/wireshark and watch the remote file operations as they happen... NFS doesn't access files by filename, but by NFS filehandle (basically device+inode number), so a remote client first looks up the filename to get the filehandle, and all accesses are done via the filehandle at that point. Theoretically, one could write a dtrace script that watches calls to nfs_namei, nfsrv_read, and nfsrv_write, and then matches read/write ops with the filenames that were looked up beforehand. Solaris NFS has a logging option, which does exactly what Sandra is asking for. It's al reason why I prefer to use Solaris for NFS servers. F. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS or an alternative?
On Jan 12, 2009, at 9:56 PM, Jay Hall wrote: I am in the process of redesigning my organization's network. And, since we will be using mostly Macintosh OS X clients, I am considering using NFS. However, I will need the ability to perform user/group authentication since users may not always log in from the same PC. Essentially, each user has a home directory which only they, and possibly their secretary, needs to have access to. And, we have directories which groups of people need access to. Given the above requirements, Samba/CIFS is probably a better match for what you are doing that NFS would be. From the reading I have done this evening, my understanding is NFSv4 will meet all of these needs. Is this correct? And, is there a better way to accomplish this? Note that Apple only ships NFSv3-aware software, and I'm not sure whether FreeBSD supports NFSv4 yet either. There appears to be external work here: http://snowhite.cis.uoguelph.ca/nfsv4/ http://www.citi.umich.edu/projects/nfsv4/ ...which you might look into. Regards, -- -Chuck ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS or an alternative?
Chuck Swiger wrote: On Jan 12, 2009, at 9:56 PM, Jay Hall wrote: I am in the process of redesigning my organization's network. And, since we will be using mostly Macintosh OS X clients, I am considering using NFS. However, I will need the ability to perform user/group authentication since users may not always log in from the same PC. Essentially, each user has a home directory which only they, and possibly their secretary, needs to have access to. And, we have directories which groups of people need access to. Given the above requirements, Samba/CIFS is probably a better match for what you are doing that NFS would be. you could try webdav. apple's iDisk. i have used this on our corporate network for a while now, and allows mounting from any workstation. From the reading I have done this evening, my understanding is NFSv4 will meet all of these needs. Is this correct? And, is there a better way to accomplish this? Note that Apple only ships NFSv3-aware software, and I'm not sure whether FreeBSD supports NFSv4 yet either. There appears to be external work here: http://snowhite.cis.uoguelph.ca/nfsv4/ http://www.citi.umich.edu/projects/nfsv4/ ...which you might look into. Regards, ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS fstab style
Can someone confirm that these two lines are the same -or- if one is preferred over the other ? Code: 192.168.1.8:/temp/tmp_nfs nfs rw,-b,-i 0 0 192.168.1.8:/temp/tmp_nfs nfs rw,bg,intr 0 0 I've never seen the style of line 1 before, no idea whether it would work or not. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS or an alternative?
On Mon, Jan 12, 2009, Jay Hall wrote: I am in the process of redesigning my organization's network. And, since we will be using mostly Macintosh OS X clients, I am considering using NFS. However, I will need the ability to perform user/group authentication since users may not always log in from the same PC. Essentially, each user has a home directory which only they, and possibly their secretary, needs to have access to. And, we have directories which groups of people need access to. From the reading I have done this evening, my understanding is NFSv4 will meet all of these needs. Is this correct? And, is there a better way to accomplish this? NFS is only part of the problem. We have done this using OS X, Linux, and FreeBSD clients using openldap for authentication and the amd automounter to handle home directories when there are multiple machines on which user's home directories may be found. When we create the openldap records, we map /home/username to /homes/username to avoid conflict with client machine's local user's directories. We have one system with about 10,000 users with multiple client machines handling mail delivery, pop, and imap to user's Maildir stores with the NFS mounted $HOME directories which has been working without a hitch for several years. In this case the main systems /home directory is NFS mounted to /homes on the client machine, specifying the tcp protocol for maximum reliability. Bill -- INTERNET: b...@celestial.com Bill Campbell; Celestial Software LLC URL: http://www.celestial.com/ PO Box 820; 6641 E. Mercer Way Voice: (206) 236-1676 Mercer Island, WA 98040-0820 Fax:(206) 232-9186 Permit me to issue and control the money of a nation, and I care not who makes its laws. -- Mayer Amschel Rothschild ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to freebsd-questions-unsubscr...@freebsd.org
Re: NFS Help
If your server is nfsv4 (freebsd 7.x) the nfs protocol used is tcp. The older machines (freebsd 5 or 6) the nfs prococol is udp... try to use the -T switch (mount_nfs option...) on the older machines so they will use tcp... hope this will help Sergio ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: NFS Help
Victor Farah wrote: Hello I have about 10 machines that are NFS clients, 5 are new and 5 are older. Anyway the new machine mount from the NFS server just fine. The older machines mount; and I can ls /mnt/data/; BUT when I ls /mnt/data/sc/ on the older machines this happens: nfs server 192.168.10.162:/data: not responding nfs server 192.168.10.162:/data: not responding nfs server 192.168.10.162:/data: not responding nfs server 192.168.10.162:/data: not responding But on the new machines they work perfectly fine? As well the old machines mount it as i stated before I can even ls the parent directory /mnt/data/ and it shows me all the directories on the mount, but anytime I ls or do anything inside there it does that or freezes. Any idea's? How old are these old systems? Do you have any ISA type ethernet cards? Read Handbook's section 30.3.6: http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/network-nfs.html I actually had this kind of trouble once, and it was due to an ISA network card. I doubt you are really using an ISA card in a production system, but some of the remedies described in the section may give you a hint of what is going on. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: NFS Help
AWESOME this worked like a charm, I added -r=1024 to it and BAM works! :) Thank you everybody! Manolis Kiagias wrote: Victor Farah wrote: Hello I have about 10 machines that are NFS clients, 5 are new and 5 are older. Anyway the new machine mount from the NFS server just fine. The older machines mount; and I can ls /mnt/data/; BUT when I ls /mnt/data/sc/ on the older machines this happens: nfs server 192.168.10.162:/data: not responding nfs server 192.168.10.162:/data: not responding nfs server 192.168.10.162:/data: not responding nfs server 192.168.10.162:/data: not responding But on the new machines they work perfectly fine? As well the old machines mount it as i stated before I can even ls the parent directory /mnt/data/ and it shows me all the directories on the mount, but anytime I ls or do anything inside there it does that or freezes. Any idea's? How old are these old systems? Do you have any ISA type ethernet cards? Read Handbook's section 30.3.6: http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/network-nfs.html I actually had this kind of trouble once, and it was due to an ISA network card. I doubt you are really using an ISA card in a production system, but some of the remedies described in the section may give you a hint of what is going on. -- Victor Farah - Systems Administrator netmediaservices.net ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: NFS Help
Sorry I forgot to mention, all machines are FREEBSD6.3 Sergio de Almeida Lenzi wrote: If your server is nfsv4 (freebsd 7.x) the nfs protocol used is tcp. The older machines (freebsd 5 or 6) the nfs prococol is udp... try to use the -T switch (mount_nfs option...) on the older machines so they will use tcp... hope this will help Sergio ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] -- Victor Farah - Systems Administrator netmediaservices.net ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: NFS drops with em0 driver
Thanks everyone for the replies so far. I have disabled ACPI on my boxes. However the drops still persist. I would be grateful if someone can provide more ideas. Thanks Subhro On Sun, Jun 1, 2008 at 2:17 PM, Catalin Miclaus [EMAIL PROTECTED] wrote: -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Subhro Sent: Sunday, June 01, 2008 7:06 AM To: freebsd-questions@freebsd.org Subject: NFS drops with em0 driver Hello, I am facing a strange problem on my systems. I am running FreeBSD-6.2-RELEASE-p12. My network interface uses the em driver. I am facing a lot of issues where the NFS connections are dying randomly. Is there any known bug with the em driver? I am using the SCHED_ULE scheduler. Thanks Subhro Check this: http://wiki.freebsd.org/JeremyChadwick/Commonly_reported_issues Best Regards Catalin Miclaus Network/Security ISP-Data Starcomms Ltd. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] -- Subhro Kar Software Engineer Dynamic Digital Technologies Pvt. Ltd. EPY-3, Sector: V Salt Lake City 700091 India ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: NFS drops with em0 driver
Subhro Hello Subhro, I have nothing to add besides a me too. I ended up replacing my em0 with an fxp0 and all of my networking issues (watchdog timeouts, link dropping/reconnection, etc.) disappeared. This may have been flaky hardware. Do you have another NIC you can try with? i use em network on FreeBSD 6.3/amd64, normal sheduler, all works fine. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: NFS drops with em0 driver
On Sun, Jun 01, 2008 at 11:36:05AM +0530, Subhro wrote: Hello, I am facing a strange problem on my systems. I am running FreeBSD-6.2-RELEASE-p12. My network interface uses the em driver. I am facing a lot of issues where the NFS connections are dying randomly. Is there any known bug with the em driver? I am using the SCHED_ULE scheduler. Thanks Subhro Hello Subhro, I have nothing to add besides a me too. I ended up replacing my em0 with an fxp0 and all of my networking issues (watchdog timeouts, link dropping/reconnection, etc.) disappeared. This may have been flaky hardware. Do you have another NIC you can try with? Thanks, Josh -- Josh Tolbert [EMAIL PROTECTED] || http://www.puresimplicity.net/~hemi/ Security is mostly a superstition. It does not exist in nature, nor do the children of men as a whole experience it. Avoiding danger is no safer in the long run than outright exposure. Life is either a daring adventure, or nothing. -- Helen Keller ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
