similar to: poor samba performance with many smaller files

Hi, I need to to open existing readonly files. Yes you are right, not the open call takes the time, but stat() system call. I looked at aio_pthread source, the lines which needs to be removed to make it work are just the lines which tests for O_CREAT|O_EXCL? What problems can cause such solution for readonly share? On Tue, Apr 26, 2016 at 8:42 PM, Jeremy Allison <jra at samba.org> wrote:

Hi, I did use wireshark to verify that the client (window7) is not the problem, client is sending out multiple "SMB2 Create request" packets before he gets some answer. On the server side samba is processing this requests in a synchronous way, in order which they were send, that is what I see from strace output. On Wed, Apr 27, 2016 at 8:31 AM, Volker Lendecke <Volker.Lendecke at

I removed the O_CREAT|O_EXCL checks and added check to exclude O_DIRECTORY, but it is not working, opening of files fails and I don't know why. I can see in strace logs that openat syscall in vfs_aio_pthread was successful, but it still fails and this messages are logged: [2016/04/28 13:28:59.691782, 5, pid=5623, effective(32697, 10513), real(32697, 0)]

The directory is small, it contains only one file. stat() takes 15ms and if I read 1000 files it sums to 15 seconds. So the issue is really that open file is synchronous. On Tue, Apr 26, 2016 at 9:57 PM, Jeremy Allison <jra at samba.org> wrote: > On Tue, Apr 26, 2016 at 09:44:57PM +0200, Peter Bulin wrote: > > Hi, > > I need to to open existing readonly files. Yes you are

On Tue, Apr 26, 2016 at 03:49:44PM +0200, Peter Bulin wrote: > Hi, > > I have problem with my samba. I have shared folder which contains many > smaller files (few MB each), this files are stored on very fast disk array. > Locally, from that machine I can read this files in parallel with speed > about 500-1000 MB/s, but when I read them through samba, I get only 50-100 > MB/s

I'm trying to learn more about samba by experimenting with samba on FreeNAS. This involves a certain amount of reading how default smb4.conf parameters are set in FreeNAS and then reading the manpages for those parameters. Async I/O is implemented in samba in freenas via the aio_pthread VFS module. The manpage for vfs_aio_pthread states: "the smb.conf parameters aio read size and aio

Dear all, After 4 days of sleepless nights, I have manged to rebuild the samba farm. I believe the following discovery might interest our samba community. ------------------------------------------------ System setting: I have deployed samba 4.1.0 system for my working organisation. It comprised of 2 DCs and 1 member server. 2 DCs maintains AD for login and the member server host files for

Hi all, I fear I can't rely on "open (path, O_CREAT | O_EXCL)" behaviour on samba share. The libc's doc says: If both `O_CREAT' and `O_EXCL' are set, then `open' fails if the specified file already exists. This is guaranteed to never clobber an existing file. But this isn't probably true on network filesystems. When I run this tiny program on

Hi Slight problem - linux 2.4 running dovecot, Solaris 2.8 home directory server. Full lockd support etc. I had to make the following code patches (at end of mail for clarity) to get dovecot to read the users mbox files at all. 1) fcntl with F_SETLKW will not work against a Solaris 2.8 server as proved with a small test program. Other forms of fcntl (ie F_SETLK) are OK. I noticed there are

Today, rsyncd manages its pid file by open()ing it with O_CREAT|O_EXCL at startup, and then unlink()ing it at shutdown. If the open() fails at startup because the file already exists, then rsyncd will assume another instance of itself is already running and not start. However, there's a problem with this approach: if rsyncd is terminated without being able to clean up (e.g., kill -9, or the

Hello, when I do: /opt/rsync/bin/rsync /etc/hosts targethost::bkp/ I get: cannot create .hosts.b0WX1x : File exists I check the targethost and I get empty file .hosts.b0WX1x When trying with other targethost-s it works, but on this one it doesn't. On the other targethosts I have exactly the same LinuX distribution, permissions and users than on the problem targethost. I have exactly the

Hi, on AIX, mkstemp doesn't open a file with the O_LARGEFILE option, so you can't transfer files > 2GB to an AIX machine. Here is a fix: diff -c -r rsync-2.5.6.orig/syscall.c rsync-2.5.6/syscall.c *** rsync-2.5.6.orig/syscall.c Sun Jan 26 21:09:02 2003 --- rsync-2.5.6/syscall.c Wed May 14 13:55:15 2003 *************** *** 151,157 **** if (dry_run) return -1; if

I have attached a patch that supports a new "--direct-write" option. The result of using this option is to write directly to the destination files, instead of a temporary file first. The reason this patch is needed is for rsyncing to a device where the device is full or nearly full. Say that I am writing to a device that has 1 Meg free, and a 2 meg file on that device is out of date.

Dear List, I'm trying to get diff/removed data and it's offset out. So I write a functions in receive_data. When I run backup, I found there is a weird behavior which I don't understand. i = recv_token(f_in, &data) will receive (i = -1, offset2 = 0) some where in the middle of the transfer procedure. That's to say, it's going to transfer the first data block from sender,

While doing migration, sometimes found suspend lock file was not unlinked in previous operations, then there is an obsolete lock file in place, which causes the current and later migration cannot get lock. That happens seldomly but do happen. After checking the source code, I found there are some places that potentially cause lock file unlinked, including: 1) in lock_suspend_event() function,

Is there an alternative to using link(2) to rename files in sftp-server? Some users use sftp to upload files to a vfat partition on an sftp-server, and then renaming doesn't work. This breaks konqueror, for example (from KDE, which u), which upload files first with a ".part" extension and then renames them removing this extension.

I found something interesting via strace. lda is writing a timestamp with utime before doign the fsync, but I'm really not a C guy, so I have no idea why that's going on via procmail and not via commandline. I assume it's related to the choice of pread64 vs read. when called from commandline (working): read(0, "July 14-20, 2013\n10 courses. Bon"..., 4096) = 4096

Hi, I noticed that Markus has fixed the temporary file cleanup problems in OpenSSH cvs. What files need patching for this ? I only noticed changes in: session.c, channels.h and channels.c. -Jarno -- Jarno Huuskonen <Jarno.Huuskonen at uku.fi>

Hi all, I am using samba 4.10.7 and it seems to have bug for using @group in valid or invalid conf (?). And i can't find fixed patch in later release. I describe this issue detail below: 1. Firstly, there is my samba conf below (Add @d_group in "invalid users"): (smb_share.conf) [f1] path = /home/f1 write list = "admin" "@Administrator_Group"

Hi list, I have 3 Citrix Xenserver 5.6  and currently building a a VM backup server. I took the xe client on one of the server and installed it on the backup server (gentoo based) installed the stunnel package to make XE work. now when running, I get this error message : [server ­~ #] ./xe help -debug -s 192.168.111.17 -u user -pw pass Connecting via stunnel to [192.168.111.17] port [443]