similar to: Nagle & delayed ACK strike again

In certain situations sftp download speed can be much less than that of scp. After many days of trying to find the cause finally I found it to be the tcp nagle algorithm, which if turned off with TCP_NODELAY eliminates the problem. Now I see it being discussed back in 2002, but it still unresolved in openssh-4.2 :( Simple solution would be to add a NoDelay option to ssh which sftp would set.

> On Dec 3, 2007 5:19 AM, Miklos Szeredi <miklos at szeredi.hu> wrote: > > > Yes it is weird. I hope you can help me with this problem as sshfs is > > > so much simpler than nfs. > > > > Can you please do one more thing: start up sshfs normally, then on the > > server do a strace of the sftp-server process: > > > > strace -o /tmp/strace.log

https://bugzilla.mindrot.org/show_bug.cgi?id=1555 Summary: add hard link and attribute extensions to sftp-server Product: Portable OpenSSH Version: 5.1p1 Platform: All OS/Version: All Status: NEW Severity: enhancement Priority: P2 Component: sftp-server AssignedTo: unassigned-bugs at mindrot.org

https://bugzilla.mindrot.org/show_bug.cgi?id=1399 Summary: add statfs extension to sftp-server Classification: Unclassified Product: Portable OpenSSH Version: -current Platform: All OS/Version: All Status: NEW Severity: normal Priority: P2 Component: sftp-server AssignedTo: bitbucket at

https://bugzilla.mindrot.org/show_bug.cgi?id=1400 Summary: add atomic rename extension to sftp-server Classification: Unclassified Product: Portable OpenSSH Version: -current Platform: All OS/Version: All Status: NEW Severity: normal Priority: P2 Component: sftp-server AssignedTo: bitbucket at

Hello again! Since there was some resistance against adding TCP_NODELAY uncontionally, I've made another patch. The new patch contains the following: * Added a NoDelay yes/no (default no) config option to ssh and sshd * Added -oNoDelay=yes to the ssh command line for sftp. * Changed the sshd subsystem config option syntax from Subsystem name path to Subsystem name options path

Dear all, I am looking for the status on the hardlink patches that were published on this list last february by Miklos Szeredi? I'd really like to have hardlink support in sftp. This would in turn enable hardlinks in sshfs and make incremental rsync backups to remote filesystems possible. can someone tell me if these patches will be incorporated in openssh? The patches can be found under

I have a system in place where it appears that TCP will make a massive change in behavior mid-stream with existing SSH sessions. We noticed the issue first with an application using an SSH forward. However, we were able to rule that out by generating the same TCP characteristics by having a perl script dump text out to a terminal simulating a large data flow from the far end(ssh server) back

On Fri, Sep 06, 2019 at 01:52:41PM +0200, Miklos Szeredi wrote: > On Fri, Sep 6, 2019 at 12:36 PM Stefan Hajnoczi <stefanha at redhat.com> wrote: > > > > On Fri, Sep 06, 2019 at 10:15:14AM +0200, Miklos Szeredi wrote: > > > On Thu, Sep 5, 2019 at 9:49 PM Vivek Goyal <vgoyal at redhat.com> wrote: > > > > > > > > Hi, > > > >

I: Program returns random data in a function E: libguestfs no-return-in-nonvoid-function guestmount.c:75 The function fuse_opt_add_opt_escaped has only one caller and a return code is not checked. Signed-off-by: Olaf Hering <olaf at aepfle.de> --- diff --git a/fuse/guestmount.c b/fuse/guestmount.c index 17e94ba..1eb0553 100644 --- a/fuse/guestmount.c +++ b/fuse/guestmount.c @@ -49,7 +49,7

On Fri, Sep 06, 2019 at 01:52:41PM +0200, Miklos Szeredi wrote: > On Fri, Sep 6, 2019 at 12:36 PM Stefan Hajnoczi <stefanha at redhat.com> wrote: > > > > On Fri, Sep 06, 2019 at 10:15:14AM +0200, Miklos Szeredi wrote: > > > On Thu, Sep 5, 2019 at 9:49 PM Vivek Goyal <vgoyal at redhat.com> wrote: > > > > > > > > Hi, > > > >

On Sun, Sep 08, 2019 at 07:53:55PM +0800, piaojun wrote: > > > On 2019/9/6 19:52, Miklos Szeredi wrote: > > On Fri, Sep 6, 2019 at 12:36 PM Stefan Hajnoczi <stefanha at redhat.com> wrote: > >> > >> On Fri, Sep 06, 2019 at 10:15:14AM +0200, Miklos Szeredi wrote: > >>> On Thu, Sep 5, 2019 at 9:49 PM Vivek Goyal <vgoyal at redhat.com> wrote:

And while we are at it, can you please comment on these patches as well, originally submitted around one year ago. Oh, and I think we can agree, that the secsh-filexfer standardization is dead, so there's not much point in trying to support newer protocol versions, which don't have statfs anyway. Thanks, Miklos ----- This is needed to be able to support statfs operation on an SSH

I have a distributed/replicated setup with Glusterfs 3.0.2, that I'm testing on 4 servers, each with access to /mnt/gluster (which consists of all directories /mnt/data01 - data24) on each server. I'm using configs I built from volgen, but every time I access a file (via an 'ls -l') for the first time, I get all of these messages in my logs on each server: [2010-05-04 10:50:30] W

Here's a patch that adds support for the creation of hard links over SFTP. Hard links are not used very often nowdays, but they do still have their uses and this is currently the most often requested improvement for SSHFS. To detect hard links the st_nlink, st_dev and st_ino attributes are usually used. I'll also post patches adding extensions for these and other attributes. Please

On Tue, Jun 23, 2020 at 5:17 PM Aki Tuomi <aki.tuomi at open-xchange.com> wrote: > > > On 23/06/2020 12:15 Joe Wong <joewong99 at gmail.com> wrote: > > > > > > Hello, I have added 'y' as nodelay in my passdb query and I see the > following in log: > > > > dovecot: imap-login: Debug: Ignoring unknown passdb extra field: nodelay >

Hello, I have added 'y' as nodelay in my passdb query and I see the following in log: dovecot: imap-login: Debug: Ignoring unknown passdb extra field: nodelay does it mean this option is no longer supported in 2.3.10 ? - Joe -------------- next part -------------- An HTML attachment was scrubbed... URL:

Hi All, I noticed a very weird phenomenon when I'm copying data (200KB image files) to our glusterfs storage. When I run only run client, it copies roughly 20 files per second and as soon as I start a second client on another machine, the copy rate of the first client immediately degrade to 5 files per second. When I stop the second client, the first client will immediately speed up

> As a streaming server, it's fairly crucial for icecast to > send out data with as low a delay as possible (many clients > don't care, but some do). That's why we use TCP_NODELAY - we > actually WANT to send out data as soon as we can. Nagle is inherently unsuited for streams. NODELAY was (imho) ment for connections for which Nagle isn't sufficient and CORK is not

On Thu, Oct 03, 2019 at 10:42:44AM +0200, Miklos Szeredi wrote: > On Wed, Oct 2, 2019 at 3:27 PM Vivek Goyal <vgoyal at redhat.com> wrote: > > > > On Wed, Oct 02, 2019 at 09:40:11AM +0200, Miklos Szeredi wrote: > > > Looking at the ugly retry logic in virtiofs and have some questions. > > > > Hi Miklos, > > > > What are you thinking w.r.t