SOLVED Slow SSH Login

[Server Pros]

I have just had my first cPanel installation installed after 18 years with Plesk. I have cPanel installed in a Virtuozzo container. It takes 30 seconds for SSH to login. I am using Absolute Telnet (SSH2 mode) that I've used and upgraded for years. The management company cannot answer the question as to why.

Any thoughts? I'm not going back to the ridiculously unhelpful Plesk products and staff. But I don't have this issue with Plesk that has been installed in another container. Instant login. Just not on the hardened cPanel.

Thank you in advance. I'm new to cPanel and may ask a stupid question here or there.

 

[24x7server]

Hi,

Basically, SSH service is not controlled by the cPanel or the Plesk. If the SSH is slow, it directly relates to your VPS/container network connectivity. This is not a cPanel issue to be precise, nor would it be a Plesk issue if Plesk was installed on it..

 

[Server Pros]

Thank you for your reply.

To be clear the problem is not that SSH is slow once logged in. It is slow to login. 30 seconds. I've tried direct and using my cellphone app over my cellphone provider. Different IP's. No firewall. I'll open a ticket at Virtuozzo. As you noted I wrote that Plesk was not slow at all so I'm not sure if this is a hardening problem. Plesk hardened itself to some extent. The cPanel installation was done by a management company near you.

 

[Server Pros]

I need to re-address this question for clarity.

1. Plesk was installed in a Virtuozzo 7 container. Access via SSH was instant.

2. We dropped Plesk like a hot potato. No need to explain why.

3. We reinstalled a new container with absolutely no changes. Same OS. Same memory, same hostname that stayed active in DNS during the switch, same IP's, etc.

4. Now that cPanel is installed it takes 30 seconds to login.

I understand that cPanel is not the OS. After 18 years as a host I am fully aware. But the container had "cPanel hardening" applied to it. What in that process could be drastically slowing SSH access?

 

[rarod]

Try this: linux-tips.com/t/disabling-reverse-dns-lookups-in-ssh/222

 

[Server Pros]

That was already checked but thank you. Please remember that I wrote that when removing Plesk the container was deleted and recreated. This means that unless the management company changed UseDNS to no since the OS was installed fresh before cPanel was installed with in it, it would not be the culprit. The OS installs the same each time and doesn't know that it is getting Plesk, cPanel, DirectAdmin or nothing at all.

This is not the only container on the hardware node. ALL OTHERS with Plesk respond instantly. So this begs the question of what is different? If not cPanel what hardening technique besides UseDNS being turned to yes and resolving hostnames could be the cuplprit. A virtuozzo container installs the OS just like a stand-alone server.

The linked to article (that I'd already read) it suggests adding the client side hostname to /etc/hosts. I'm using Windows on a shared static IP and I'm not aware that my laptop has a hostname like a linux server does, so add what to /etc/hosts?

 

[cPanelLauren]

Hi @easyprosys

I've had one other report of this that I can recall. The thread here SOLVED - Delayed response on ssh connection the user found the issue and let us know the resolution that worked for him, can you let us know if it works for you as well.

Thanks!

 

[garconcn]

I've seen this issue quite frequent on CentOS/Cloudlinux 7 servers but not on version 6.

First, check systemd-logind status

# systemctl status systemd-logind

If you see error like below, restart the systemd-logind service will fix the issue:

systemd-logind[13432]: Failed to abandon session scope: Transport endpoint is not connected

Restart it

# systemctl restart systemd-logind

 

[Server Pros]

I did NOT find that error but "Failed to abandon session scope: Transport endpoint is not connected" WAS I restarted anyway it and it worked! THANK YOU! EXCELLENT! Let's hope it doesn't recur. If it does I'll have the fix. Do you know if this is in the cPanel KB? I didn't find it and maybe should add it (suggest.)

 

[Server Pros]

THIS IS SOLVED.

 

[garconcn]

My reply seems to have disappeared.The error code was not present but "Failed to abandon session scope: Transport endpoint is not connected" was. I restarted anyway it and it worked! Let's hope it doesn't recur.

THIS IS SOLVED.

Do you mean this error code "systemd-logind[13432]"?
I think it's process ID, it should be different from yours, and every time you restart the service, it will change.

I found this issue happen like once a month or two months. I don't know why it happened and how to fix it permanently.

 

[Server Pros]

The error code was not there at all. But the phrase "Failed to abandon session scope: Transport endpoint is not connected" was. I restarted it and it worked. I've since read that this is a known condition with Debian, Ubuntu and now we see CentOS 7.

https://www.linuxquestions.org/questions/debian-26/systemd-logind-slow-logins-4175543760/

https://ubuntuforums.org/showthread.php?t=2327694

Thank you ALL for your help. Not a cPanel problem but for some reason did not occur when another CP was installed. We'll see if I need to add this to rc.local.

 

[cPanelLauren]

Hi @easyprosys

I'm really happy to hear that you were able to resolve the issue! Thank you for letting us know as well! I don't believe this is in a cPanel related kb mostly because it's specific to CentOS7.

It seems to be a bit random in regard to which servers experience an issue personally I haven't had any of my test servers experience it though you are the second person I've talked to who has.

I'm going to go ahead and mark this as resolved at this time but I want to tell you we're happy you chose cPanel and please don't hesitate to ask if you have any other questions!

Thanks!

 

[Server Pros]

Hi @easyprosys

I'm really happy to hear that you were able to resolve the issue! Thank you for letting us know as well! I don't believe this is in a cPanel related kb mostly because it's specific to CentOS7.

It seems to be a bit random in regard to which servers experience an issue personally I haven't had any of my test servers experience it though you are the second person I've talked to who has.

I'm going to go ahead and mark this as resolved at this time but I want to tell you we're happy you chose cPanel and please don't hesitate to ask if you have any other questions!

Thanks!

Actually, it isn't just CentOS 6 or 7. It is also at least Debian and Ubuntu. I posted the links above. There may be others. As I understand it a yum update is done after installing cPanel. The container exhibited no issue until after installing cPanel and that somewhat suggests that something with the installation and/or hardening of the server are the culprit. Since Plesk Onyx had been installed and a yum update done and no problem occurred it makes one wonder. I am not implying anything but rather passing on details that may help someone else in the future.

Regardless of which is at fault I'd recommend that this issue be added to the KB since several OS distributions cause the issue to occur.

 

[cPanelLauren]

HI @easyprosys

Actually, it isn't just CentOS 6 or 7. It is also at least Debian and Ubuntu.

You're correct but the issue being present on Debian and Ubuntu is not relevant to cPanel since we only support CentOS 6 & 7.

As I understand it a yum update is done after installing cPanel.

A yum update is done, this is correct but no matter what you do, a yum update wouldn't be something which would only run upon installing cPanel.

Since Plesk Onyx had been installed and a yum update done and no problem occurred it makes one wonder. I am not implying anything but rather passing on details that may help someone else in the future.

I have several test servers none of which have experienced the issue, just because you didn't experience it when using another panel does not make it something specific to cPanel. I understand your concern though and this post will remain to help others who may be experiencing the issue.

Thanks!

 

[Server Pros]

HI @easyprosys


You're correct but the issue being present on Debian and Ubuntu is not relevant to cPanel since we only support CentOS 6 & 7.


A yum update is done, this is correct but no matter what you do, a yum update wouldn't be something which would only run upon installing cPanel.



I have several test servers none of which have experienced the issue, just because you didn't experience it when using another panel does not make it something specific to cPanel. I understand your concern though and this post will remain to help others who may be experiencing the issue.

Thanks!

Thank you for teaching me something about cPanel. I didn't know that it only ran on CentOS! I agree that the issue is likely OS related but only say that it did not occur until after cPanel was installed. It is a Virtuozzo 7.4 installation that allowed access without a delay. Maybe it has something to do with the hardening. We'll see what happens when I install the second cP in a new container.

Thanks again.