Error 1831-008 when mounting a NFS share from Linux server to AIX machine
Oliver J Lindner
I'm having problems mounting a NFS share from a Linux server to an IBM
RS/6000 model 43P running AIX 4.3.
The mount command "mount LinuxBox:/SomeNfsShare /ProperAixMountPoint"
fails with a mount/vmount 1831-008 error messages (giving up on
DeviceName).
The AIX Message Guide and Refernce was no help. The suggested recovery
method "reboot" does not help, as does the "NFS Problem Determination"
section in the System Management Guide.
- Both machines can ping each other, or telnet.
- NFS shares from the AIX machine to the Linux box are also possible.
- All NFS shares from the Linux machine can be seen using the
"showmount -e" command.
- All Linux shares can also be used by other (non-IBM) machines on the
net.
- The inetd, portmap, and nfs deamons are running on the AIX and on
Linux.
- A valid mount point on the AIX exists, i.e. I can mount other
devices to this mount point properly.
- Running the "/usr/bin/rpcinfo -p Linux" command returns alls
reuqired information stanzas.
Any ideas? Regards. <Oliver>
Mark
likely it is. Run "nfso -a" and see what 'nfs_use_reserved_ports' is set
to. If it's set to '0', then that's your problem. Run:
# nfso -o nfs_use_reserved_ports=1
This should solve your problem. Be sure to place this command in
/etc/rc.net, so that it survives reboots. Let me know if this works.
Dawson
removenosp...@odc.net
: Is the AIX box attempting the mount from an unprivileged port? Most
: likely it is. Run "nfso -a" and see what 'nfs_use_reserved_ports' is set
: to. If it's set to '0', then that's your problem. Run:
: # nfso -o nfs_use_reserved_ports=1
: This should solve your problem. Be sure to place this command in
: /etc/rc.net, so that it survives reboots. Let me know if this works.
: Dawson
I have the exact same problem, it only happens on AIX 4.3.X systems, not
4.1 or other clients we have here...
I tried the nfso -o nfs_use_reserved_ports=1
also on /etc/rc.nfs and made sure it was set to use by doing nfso -a
Same problem..
Must be a bug with the AIX nfs client..
Oliver J Lindner
>Is the AIX box attempting the mount from an unprivileged port? Most
>likely it is. Run "nfso -a" and see what 'nfs_use_reserved_ports' is set
>to. If it's set to '0', then that's your problem. Run:
>
># nfso -o nfs_use_reserved_ports=1
>
>This should solve your problem. Be sure to place this command in
>/etc/rc.net, so that it survives reboots. Let me know if this works.
Yes, it works. Setting the nfs option to 1 will allow the mount
command to process correctly.
The "performance" for file operations on the mounted share is extremly
weak and I receive plenty of timeout errors ("A remote host did not
respond within the timeout period").
Timeout errors occur regardless of number of biod/nfsd deamons used,
whether Linux box is busy, size of file. The showmount and mount
command will work properly, and I can also use "ls" commands etc on
the mounted share.
Is there any other requirement to connect AIX and Linux machines?
<Oliver>
Mark
Have you checked "netstat -s" for udp buffer overflows?
What about "nfsstat -c" for the clients and "nfsstat -s" for servers?
This info can help determine exactly what you need to tune. By the
way, the attributes to tune can be found in "no -a" and "nfso -a."
Dawson
Oliver J Lindner
>That's great that it worked. But now onto the next problem.
>Have you checked "netstat -s" for udp buffer overflows?
>What about "nfsstat -c" for the clients and "nfsstat -s" for servers?
>This info can help determine exactly what you need to tune. By the
>way, the attributes to tune can be found in "no -a" and "nfso -a."
Since I am not very goot at analysing the output of no, nfso, et. al,
I have the results attached to this message. Maybe you could find out
whats wrong with my setup from these files.
Using the "nfs_use_reserved_port" setting I can mount the Linus NFS
share. I can also use "ls" and copy small files to/from the share now.
Copying/moving larger files (i.e. size > 4 kB) results in the
following message:
cp: <FILENAME>: A remote host did not respond
within the timeout period.
Also, I receive plenty of "permission denied" errors when trying to
create files on the share, although users and groups have identical
UID/GID on all machines.
Thanks. <Oliver>
Mark
# chdev -l ent0 -a tx_que_size=150 -a rx_que_size=150
Since you have at least 256MB of RAM on each box, set sb_max to the value
that I sent to you via email. Then, since we're dealing with udp packets, in-
crease the udp_sendspace and udp_recvspace to 131072(Your socket sizes
buffer sizes are still at the default, which is WAY TOO SMALL):
# no -o udp_sendspace=131072
# no -o udp_recvspace=131072
Also, set nfs_socketsize to 131072:
# nfso -o rfc1323=1
# nfso -o nfs_socketsize=131072
# stopsrc -s nfsd
# startsrc -s nfsd
This should help a great deal, since you've already tuned the nfsd and biod
threads amount. Let me know how it goes.
Dawson
Oliver J Lindner wrote:
> --------------------------------------------------------------------------------
> OUTPPUT OF MOUNT COMMAND ON AIX:
> --------------------------------------------------------------------------------
>
> node mounted mounted over vfs date options
> -------- --------------- --------------- ------ ------------ ---------------
> linuxserver /share /mountpoint nfs Dec 31 12:46 bg,soft,intr,rw
>
> --------------------------------------------------------------------------------
> OUTPPUT OF NO COMMAND ON AIX:
> --------------------------------------------------------------------------------
>
> extendednetstats = 0
> thewall = 262116
> sockthresh = 85
> sb_max = 1048576
> somaxconn = 1024
> clean_partial_conns = 0
> net_malloc_police = 0
> rto_low = 1
> rto_high = 64
> rto_limit = 7
> rto_length = 13
> inet_stack_size = 16
> arptab_bsiz = 7
> arptab_nb = 25
> tcp_ndebug = 100
> ifsize = 8
> arpqsize = 1
> ndpqsize = 50
> route_expire = 0
> send_file_duration = 300
> fasttimo = 200
> routerevalidate = 0
> nbc_limit = 0
> nbc_max_cache = 131072
> nbc_min_cache = 1
> strmsgsz = 0
> strctlsz = 1024
> nstrpush = 8
> strthresh = 85
> psetimers = 20
> psebufcalls = 20
> strturncnt = 15
> pseintrstack = 12288
> lowthresh = 90
> medthresh = 95
> psecache = 1
> subnetsarelocal = 1
> maxttl = 255
> ipfragttl = 60
> ipsendredirects = 1
> ipforwarding = 0
> udp_ttl = 30
> tcp_ttl = 60
> arpt_killc = 20
> tcp_sendspace = 16384
> tcp_recvspace = 16384
> udp_sendspace = 9216
> udp_recvspace = 41920
> rfc1122addrchk = 0
> nonlocsrcroute = 0
> tcp_keepintvl = 150
> tcp_keepidle = 14400
> bcastping = 0
> udpcksum = 1
> tcp_mssdflt = 512
> icmpaddressmask = 0
> tcp_keepinit = 150
> ie5_old_multicast_mapping = 0
> rfc1323 = 0
> pmtu_default_age = 10
> pmtu_rediscover_interval = 30
> udp_pmtu_discover = 0
> tcp_pmtu_discover = 0
> ipqmaxlen = 100
> directed_broadcast = 1
> ipignoreredirects = 0
> ipsrcroutesend = 1
> ipsrcrouterecv = 0
> ipsrcrouteforward = 1
> ip6srcrouteforward = 1
> ip6_defttl = 64
> ndpt_keep = 120
> ndpt_reachable = 30
> ndpt_retrans = 1
> ndpt_probe = 5
> ndpt_down = 3
> ndp_umaxtries = 3
> ndp_mmaxtries = 3
> ip6_prune = 2
> ip6forwarding = 0
> multi_homed = 1
> main_if6 = 0
> main_site6 = 0
> site6_index = 0
> maxnip6q = 20
> llsleep_timeout = 3
> tcp_timewait = 1
> tcp_ephemeral_low = 32768
> tcp_ephemeral_high = 65535
> udp_ephemeral_low = 32768
> udp_ephemeral_high = 65535
> delayack = 0
> delayackports = {}
>
> --------------------------------------------------------------------------------
> OUTPPUT OF NFSO COMMAND ON AIX:
> --------------------------------------------------------------------------------
>
> portcheck= 0
> udpchecksum= 1
> nfs_socketsize= 60000
> nfs_tcp_socketsize= 60000
> nfs_setattr_error= 0
> nfs_gather_threshold= 4096
> nfs_repeat_messages= 0
> nfs_udp_duplicate_cache_size= 5000
> nfs_tcp_duplicate_cache_size= 0
> nfs_server_base_priority= 0
> nfs_dynamic_retrans= 1
> nfs_iopace_pages= 0
> nfs_max_connections= 0
> nfs_max_threads= 8
> nfs_use_reserved_ports= 1
> nfs_device_specific_bufs= 1
> nfs_server_clread= 1
> nfs_rfc1323= 0
> nfs_max_write_size= 0
> nfs_max_read_size= 0
> nfs_allow_all_signals= 0
>
> --------------------------------------------------------------------------------
> OUTPPUT OF NETSTAT COMMAND ON AIX BEFORE MOUNTING SHARE:
> (ALL STANZAS WITH ZERO VALUE REMOVED)
> --------------------------------------------------------------------------------
>
> ip:
> 45196 total packets received
> 45194 packets for this host
> 2 packets for unknown/unsupported protocol
> 46235 packets sent from this host
> icmp:
> 2 calls to icmp_error
> Output histogram:
> destination unreachable: 2
> Input histogram:
> destination unreachable: 2
> igmp:
> 4 membership reports sent
> tcp:
> 37911 packets sent
> 22006 data packets (5973543 bytes)
> 12027 ack-only packets (8678 delayed)
> 203 window update packets
> 3675 control packets
> 38626 packets received
> 22542 acks (for 5973120 bytes)
> 709 duplicate acks
> 22581 packets (7283014 bytes) received in-sequence
> 23 completely duplicate packets (8123 bytes)
> 629 out-of-order packets (20440 bytes)
> 95 window update packets
> 931 packets received after close
> 1583 connection requests
> 551 connection accepts
> 2132 connections established (including accepts)
> 2635 connections closed (including 14 drops)
> 2 embryonic connections dropped
> 24116 segments updated rtt (of 25005 attempts)
> 6 retransmit timeouts
> 29 keepalive timeouts
> 13 keepalive probes sent
> udp:
> 6568 datagrams received
> 2 dropped due to no socket
> 1119 broadcast/multicast datagrams dropped due to no socket
> 0 socket buffer overflows
> 5447 delivered
> 5449 datagrams output
>
> ipv6:
> icmpv6:
>
> --------------------------------------------------------------------------------
> OUTPPUT OF NETSTAT COMMAND ON AIX AFTER WRITING TO MOUNT:
> (ALL STANZAS WITH ZERO VALUE REMOVED)
> --------------------------------------------------------------------------------
>
> ip:
> 47260 total packets received
> 47258 packets for this host
> 2 packets for unknown/unsupported protocol
> 48589 packets sent from this host
> icmp:
> 2 calls to icmp_error
> igmp:
> 4 membership reports sent
> tcp:
> 39706 packets sent
> 23064 data packets (6095589 bytes)
> 12676 ack-only packets (9257 delayed)
> 218 window update packets
> 3748 control packets
> 40162 packets received
> 23531 acks (for 6095192 bytes)
> 722 duplicate acks
> 23782 packets (7511659 bytes) received in-sequence
> 29 completely duplicate packets (9428 bytes)
> 636 out-of-order packets (20440 bytes)
> 100 window update packets
> 946 packets received after close
> 1621 connection requests
> 557 connection accepts
> 2176 connections established (including accepts)
> 2673 connections closed (including 14 drops)
> 2 embryonic connections dropped
> 25143 segments updated rtt (of 26047 attempts)
> 6 retransmit timeouts
> 0 connections dropped by rexmit timeout
> 29 keepalive timeouts
> 13 keepalive probes sent
> 0 connections dropped by keepalive
> udp:
> 7096 datagrams received
> 2 dropped due to no socket
> 1145 broadcast/multicast datagrams dropped due to no socket
> 5949 delivered
> 5951 datagrams output
>
> ipv6:
> icmpv6:
>
> --------------------------------------------------------------------------------
> OUTPPUT OF NFSSTAT COMMAND ON AIX BEFORE MOUNTING SHARE:
> --------------------------------------------------------------------------------
>
> Client rpc:
> Connection oriented
> calls badcalls badxids timeouts newcreds badverfs timers
> 1216 268 0 184 0 0 0
> nomem cantconn interrupts
> 0 84 0
> Connectionless
> calls badcalls retrans badxids timeouts newcreds badverfs
> 53 0 0 0 0 0 0
> timers nomem cantsend
> 0 0 0
>
> Client nfs:
> calls badcalls clgets cltoomany
> 1163 135 0 0
> Version 2: (1163 calls)
> null getattr setattr root lookup readlink read
> 0 0% 81 6% 0 0% 0 0% 619 53% 0 0% 409 35%
> wrcache write create remove rename link symlink
> 0 0% 1 0% 6 0% 1 0% 0 0% 0 0% 0 0%
> mkdir rmdir readdir statfs
> 3 0% 0 0% 26 2% 17 1%
> Version 3: (0 calls)
> null getattr setattr lookup access readlink read
> 0 0% 0 0% 0 0% 0 0% 0 0% 0 0% 0 0%
> write create mkdir symlink mknod remove rmdir
> 0 0% 0 0% 0 0% 0 0% 0 0% 0 0% 0 0%
> rename link readdir readdir+ fsstat fsinfo pathconf
> 0 0% 0 0% 0 0% 0 0% 0 0% 0 0% 0 0%
> commit
> 0 0%
>
> --------------------------------------------------------------------------------
> OUTPPUT OF NFSSTAT COMMAND ON AIX AFTER WRITING TO SHARE:
> --------------------------------------------------------------------------------
>
> Client rpc:
> Connection oriented
> calls badcalls badxids timeouts newcreds badverfs timers
> 1375 272 0 187 0 0 0
> nomem cantconn interrupts
> 0 85 0
> Connectionless
> calls badcalls retrans badxids timeouts newcreds badverfs
> 53 0 0 0 0 0 0
> timers nomem cantsend
> 0 0 0
>
> Client nfs:
> calls badcalls clgets cltoomany
> 1322 139 0 0
> Version 2: (1322 calls)
> null getattr setattr root lookup readlink read
> 0 0% 109 8% 7 0% 0 0% 709 53% 0 0% 412 31%
> wrcache write create remove rename link symlink
> 0 0% 6 0% 22 1% 1 0% 0 0% 0 0% 0 0%
> mkdir rmdir readdir statfs
> 6 0% 0 0% 32 2% 18 1%
> Version 3: (0 calls)
> null getattr setattr lookup access readlink read
> 0 0% 0 0% 0 0% 0 0% 0 0% 0 0% 0 0%
> write create mkdir symlink mknod remove rmdir
> 0 0% 0 0% 0 0% 0 0% 0 0% 0 0% 0 0%
> rename link readdir readdir+ fsstat fsinfo pathconf
> 0 0% 0 0% 0 0% 0 0% 0 0% 0 0% 0 0%
> commit
> 0 0%
removenosp...@odc.net
: share. I can also use "ls" and copy small files to/from the share now.
: Copying/moving larger files (i.e. size > 4 kB) results in the
: following message:
: cp: <FILENAME>: A remote host did not respond
Regarding this problem..
I also had the same problem using the userspace nfs server for linux..
I installed the kernel nfs server (2.1.132ac3) and the related tools,
the problem went away..
Been working great on our aix 4.3.2 boxes ever since, before that,
I had the same problem - 1831-008