• We value your experience with Plesk during 2024
    Plesk strives to perform even better in 2025. To help us improve further, please answer a few questions about your experience with Plesk Obsidian 2024.
    Please take this short survey:

    https://pt-research.typeform.com/to/AmZvSXkx
  • The Horde webmail has been deprecated. Its complete removal is scheduled for April 2025. For details and recommended actions, see the Feature and Deprecation Plan.
  • We’re working on enhancing the Monitoring feature in Plesk, and we could really use your expertise! If you’re open to sharing your experiences with server and website monitoring or providing feedback, we’d love to have a one-hour online meeting with you.

Resolved SFTP no longer accessible

Franco

Regular Pleskian
Hello,

not sure if this is due to a Plesk update or something else on the OS (CentOs 7.4), but suddenly my chrooted configured ftp accounts are no longer accepted ("Received unexpected end-of-file from SFTP server").
They've been running for years and suddenly they get blocked/rejected. I do not see anything special on the firewall.

Any idea what that might be, please?

Regards

Franco
 
Hi, I saw that, thanks, my anticipated update was a kind of mistake in fact...
Anyway, do you think it's likely due to that or shall I keep searching?
 
Hi Franco,

actually, I experienced as well some ( tiny ) issues/errors/problems with my upgrade - tests to CentOS 7.4 and I would strongly recommend to wait for the Plesk developpment and it's announcement and for possible updates/patches regarding the upgrade.
 
I just noticed that each attempt to connect to SFTP produces a segfault, such as this one:

Sep 21 16:20:43 vps kernel: bash[12928]: segfault at 1 ip 00007f28927033bc sp 00007fffc4a6bcc0 error 4 in libc.so.6[7f28925cf000+1b8000]

By this it looks to me that Plesk is not involved, most likely it's a CentOS 7.4 bug.
 
Hi Franco,

Sep 21 16:20:43 vps kernel: bash[12928]: segfault at 1 ip 00007f28927033bc sp 00007fffc4a6bcc0 error 4 in libc.so.6[7f28925cf000+1b8000]

By this it looks to me that Plesk is not involved, most likely it's a CentOS 7.4 bug.
The "libc.so.6" library is part of the "glibc" - rpm, so you are correct by stating, that this is not Plesk - related. :)
 
Plesk has now announced support for CentOS 7.4, but the libc.so.6 segfault remains and my SFTP is still unavailable :(
Am I the only one? Any suggestion?
 
I see that it is Plesk installed on VPS. Most probably that these segfaults may be caused by lack of VPS resources and calling kernel OOM killer as result. Have you checked this?
 
Hi,
I doubt that's the case. I have 400KB memory available and I see no other message indicating some issues. Perhaps file discriptors and things like that? How do I monitor them, please?
The segfautl only happens when somebody attemps an ftp access and the sequence it's always the same (the dots are to mask the username):

Sep 25 15:22:46 vps systemd: Created slice User Slice of ...
Sep 25 15:22:46 vps systemd: Starting User Slice of ...
Sep 25 15:22:46 vps systemd: Started Session 3737 of user ....
Sep 25 15:22:46 vps systemd: Starting Session 3737 of user ....
Sep 25 15:22:46 vps systemd-logind: New session 3737 of user ....
Sep 25 15:22:46 vps kernel: bash[30509]: segfault at 1 ip 00007f56fa6dd3bc sp 00007ffd32647a40 error 4 in libc.so.6[7f56fa5a9000+1b8000]
Sep 25 15:22:46 vps systemd-logind: Removed session 3737.
Sep 25 15:22:46 vps systemd: Removed slice User Slice of .....
Sep 25 15:22:46 vps systemd: Stopping User Slice of .....
 
i notices that libc.so.6 is in all vhosts folders, such as :
2127336 Aug 1 23:09 /var/www/vhosts/.../usr/lib64/libc.so.6

but in /usr/ I read:
12 Sep 15 09:04 /usr/lib64/libc.so.6 -> libc-2.17.so
2127336 Aug 1 23:09 /usr/lib64/libc-2.17.so

What does that mean? Am I running 2 different versions?
 
Hi,
for information, the error has disappeared. When trying today with filezilla I got a warning about a key change and that it could be a security threat (I lost the actual message). I accepted and continued and voila, all is there. There weren't any updated and I have not played with keys. I wonder whether that key change has anything to do with the lib crash or with the CentOS 7.4 upgrade, days after it happened.
I mark the thread as resolved, but questions remain...
 
Back
Top