similar to: Re: Filelocking-Problem: Mars_NWE together with Samba

--------------------------------------------------------------------- Red Hat, Inc. Security Advisory Synopsis: Buffer overflow in mars_nwe Advisory ID: RHSA-1999:037-01 Issue date: 1999-09-13 Updated on: Keywords: mars_nwe buffer Cross references: --------------------------------------------------------------------- 1. Topic: There are several buffer overruns in the mars_nwe

One of our group companies has just 'gone live' on a new laboratory management package. Although the software is M$-based (ie: expects a M$ server at the heart of it), it is running with its core database on a CentOS3 server using Samba for client access - in fact, the developers are amazed at the performance and say it beats a Win2K/2K3 server 'hands down'!!! Anyway, the old lab

I've now tried 2.0.6pre2 with a SuperMicro P6DNF dual PPro as well as the Everex StepDP/Pro dual PPro MB; same result, large (more than 1 MB) file copies usually crash with only a message about oplock errors in the logfile. Running a single CPU kernel works fine (pre2 seems much faster, but I haven't done any testing). 2.0.5a does the same. NFS and mars_nwe with the same hardware and

Hi, Can someone give me some pointer or links to understand how Filelocking and rquotad works on NFS ? I searched a lot on google , didn't get any good articles on that. -- Regards Basil

Hi Felix, On 4/18/24 08:33, Stolte, Felix via samba wrote: > Is this a bug or maybe a misconfiguration? In the latter, which parameters could cause this? I guess the nodes end up using different combinations of dev/inode as primary key for the locking.tdb record. Probably because the device numbers differ on the nodes and you didn't fix this known issue with GPFS with the fileid VFS

Hi folks, we are running a clustered Samba Fileserver using CTDB with two public IPs. We observed the following (mis) behavior: If user A with write permissions opens an excel file via public ip A and User B opens the same file via public ip B, both users get a Deny-Write lock on the same file and overwrite each others changes: 0:3693691 145522 DENY_WRITE 0x12019f RDWR

Using imapsync I changed many ACLs on many accounts simultaneously in four concurrent IMAP-sessions. I started with an empty file "shared-mailboxes". But after the sync the file "shared-mailboxes" has been crashed. I found broken lines and I'm missing some entrys. Is it possible that there's no filelocking on that file? I used the ext3 filesystem. That's

Hi there, is it possible and when, how - to change the filelocking-status of an opened file from EXCLUSIVE+BATCH to another status ? We have to open an file to write at the same time from different pc's into this file. For example the following smbstatus -u USER output: Pid DenyMode Access R/W Oplock Name DENY_DOS 0x1 RDONLY EXCLUSIVE+BATCH

Hi, first of all, please CC me on any reply, as I am only subscribed to the digest. OK. Here is the problem. Said kernel (from 4.4) seems to have problems with file-locking when the system is under high, likely network related, load. The symptoms are things using file locking (rpm, the user-space automounter amd) fail to obtain locks, usually stating timeout problems. The sytem in question is

I would like to add my case for a wish of IPX support in Samba: We use a number of CNC tool machines whose host is pure DOS based. Don't be surprised that DOS is stilll running somewhere. It is fairly OK when we get to the field of HW and real-time control, unlike Windows and the like... These hosts use IPX to access LAN shares. IPX is here preferable to TCP/IP because of its lower memory

I am looking for experience with migrating from Novell to Linux.I have Novell 4.2 - file and print server. I am looking for tools to migrate Novell users (and if possible printers) to Samba. Thanks and regards Max . 972-3-9188325

Hi, I've successfully got DOS booting over PXE with Memdisk, but as RAM is severely limited in the PCs I'm using (only 32-64MB) I can't load a huge Memdisk image to run large DOS programs, Windows, etc. Are there any programs out there like Memdisk, but instead of storing the disk in RAM they read it over the network? (A "netdisk" program.) This would be a really useful

Greetings ... > Has anyone ever got command line only Windows 95 DOS to connect to a Samba > machine? I can connect to an NT machine no problem using "net use ...", but > the only machine the DOS machine will see on the network is the NT machine; > all of the others, including my Linux machines and a Win 2k machine, it just > refuses to see. First, any program you

FWIW, i had the same problem (obtaining a byte-range lock with an offset of 0xeffffff0..0xefffffff) quite some time ago causing dBase IV on DOS machines with VLM clients to a mars_nwe server to fail to synchronize. The solution was probably the same as here, to fold this range into 0x7ffffffx. -- Morio Taneda admin,sociology,humanities,karlsruhe university,germany

Hi folks, I've ran many circles about this topic, read tons of documents. But have not found a working solution. Here is the scenario: There is a network contains around 350+ nodes, mixed Win98 and Win2k. There are four mars_nwe servers they use. We'd like to migrate that 4 servers into an AFS cell and offer it thru Samba which should be acting as a PDC. So far it seems not too

Hi We are using a Debian 1.3.1 server (two Ethernet-Cards) with 18 Clients. The Clients remote-boot from the server using mars_nwe(Novell-Server emulation). Under Windows (WfW 3.11) the Clients user the Microsoft 32-bit Tcp-Stack to access the samba-server( 1.9.18p7). The problem is that the clients can't see the samba-server in their browse-lists, nor can they access the server with net

Using the patches from -current (aa thorugh bb) samba 2.0.6 release compiles under FreeBSD 3.3-RELEASE. "patch-ba" needs to be edited to change the location of the lines patched, and "patch-bb" needs to be edited to remove the first part (no longer needed) and also change the line location for the remaining part. Throughput using an NT4 SP6 P6/200 client connected at 100Mbits

I am running FreeBSD 3.3-RELEASE with samba 2.0.6 compiled from sources with patches taken from 2.0.5a (not the current FreeBSD port, but I'll be testing that soon with 3.4-RELEASE). The socket options file is socket options = IPTOS_LOWDELAY TCP_NODELAY SO_SNDBUF=262144 SO_RCVBUF=262144 which is due to Martin Welk. Currently large files (1GB or greater)are transferring at up to 2.7MB/sec;

First, my installation is samba 2.2.8 running as a daemon (not inetd) on Solaris 8 configured to use a WINS server and provide authentication via a PDC (same host as WINS server). I do have an lmhosts file in the samba /lib directory with the PDC/WINS server address in it. My globals section; [global] workgroup = WORKGROUP netbios name = SAMBASERVER security =

I am not sure if your problem is similar to what we have just experienced. we are using samba 2.2.3a-12.3 on debian with winxp sp1 clients we are using roaming profiles first logon after reboot fails sortof Windows cannot find a server copy of roaming profile will logon with local profile funny thing we never saw any activity in hostname.log (seperate smb.log for each machine) smbstatus showed