This forum is closed to new posts and responses. New discussions are now taking place in the IBM Developer Answers forum.

Nov 30, 2013, 2:26 PM
8 Posts

Problems with binding smtp service to port 25

  • Category: Domino Server
  • Platform: Linux
  • Release: 9.0
  • Role: Administrator
  • Tags: bindsock,bind,smtp,problem
  • Replies: 21

I have problems running lotus domino smtp service. Service tries to bind to port 25 with suid helper bindsock.
I get an error, SMTP Server: Listener failure: 'bindsock' is missing, not executable, not owned by root, not setuid root or user needs net_privaddr privilege.

No other app is holding this port. ping resolves proper host name, nsswitch.conf has proper values for hosts: files, dns, myhostname

Any suggestions?


This happens on latest linux (kernels) install, on older installations domino works fine.


Dec 1, 2013, 10:10 AM
8 Posts
No, google is working fine, but these standard procedures do not work

I cant get services smtp, pop3, http, working, cause they use lower ports, and suid bit doesn't help.

It looks that latest kernel or linux install doesn't work like it should. I tried also with file capabilities, but it also dont work.

Any suggestions? 

Dec 1, 2013, 11:48 AM
36 Posts

selinux is off?  i dont like new kernel on my servers......

Recheck status default mail service, like postfix

Dec 1, 2013, 4:30 PM
8 Posts
selinux is not enabled, but the issue is in 3.12.* kernel

I seems that it disables setuid functionality

Dec 3, 2013, 7:11 AM
2 Posts
kernel 3.5.0-44

I have the exact same problem.  The one server on is on kernel 3.2.0-57 and it works fine.

Any solution?

Dec 12, 2013, 7:30 AM
2 Posts
Same problem on ubuntu

We have the same problem on Ubuntu 12.04. When upgrading kernel from version 3.5.0-43-generic to 3.5.0-44-generic, the domino HTTP task refuse to start with the same bindsock error message.

Right now we just boot to 3.5.0-43 but we don't see this as a permanent solution. And yes we know that Ubuntu is not a supported platform for domino :-)

Jan 2, 2014, 9:14 AM
3 Posts

...just using older Kernels for now. Anything newer than 3.5.0-43 don't like it, and as Gregor Ibic says, might be setuid. Can't seem to find anything apart from monitoring in the change log.

Tried looking into Capabilties but nothing I do seems to help. I even adding cap_setuid to bindsock :-(

Jan 4, 2014, 6:31 PM
3 Posts
Re: Problems with binding smtp service to port 25

I am using Ubuntu 13.10 and experiencing the same from a security update.

Before kernel 3.11.0-13 Domino was working fine. After the kernel 3.11.0-14, bindsock error occurs.

Jan 7, 2014, 6:08 AM
4 Posts
same problem

Had this problem too when upgrading from kernel 3.2.0-56 to 3.2.0-58 with Domino 9.0.1

Jan 14, 2014, 1:19 PM
3 Posts
Good news... for Domino on new > 3.5.0-44 Ubuntu Kernels

I raised a Bug Report and Ubuntu have a fix already. Don't know just yet when it'll be available in release.

Ubuntu support was quick and deadly accurate !!!

Jan 15, 2014, 8:32 AM
3 Posts
From the Ubuntu Kernel team....

HUUUGE thanks to the Ubuntu Kernel team for such a quick turn around, think the bug report was just 5 hours old, and the patched Kernel resolved the issue on my 12.04 3.5.0-45 AMD64 VM. Pretty impressive !!!

So we just need to keep booting into the old Kernel till the new Kernels are released. I found this the solution from "Bealer Jul 4 '13 at 16:45"  a better solution, by using the text entry rather than the numerical entry:

What they said...

Ok this patch has hit all of our upstream stable trees and will hit the
various kernels in the next SRU cycle.  Note that this is not the next
kernel which will hit the archive for all releases, for precise it is
included in that very next kernel, other releases it will be the
following kernel.

Feb 21, 2014, 5:42 PM
3 Posts
Fixed ?

I am using Ubuntu 13.10. I updated the kernel to 3.11.0-17 through aptitude and the problem seems to be gone.

May 27, 2014, 5:04 PM
3 Posts
Re: Problems with binding smtp service to port 25

This problem occurs again at kernel

I am using Ubuntu 14.04 LTS.

Kernal of the initial version of 14.04 LTS is OK.

Nov 28, 2014, 11:37 AM
46 Posts
Still a problem but- Found a workaround!

Still a problem in 3.13.0-39 and 3.13.0-40.47 BUT... I've found a workaround. First of all, this is a bug in the KERNEL. It's messing up setuid. This has been FIXED in various other lines of the kernel which is where my workaround comes from.

First- don't settle for a workaround. Log into Ubuntu's launchpad and add your voice. If the developers don't know it's not affecting lots of people it won't get much attention. The bugs to comment on are here:

  1. (this is the one that was fixed)
  2. (this is the open bug)

Next, this is the workaround I figured out for Ubuntu 14.04.1 LTS (Trusty Tahr). Until Trusty's kernels get fixed, use the working Ubuntu 12.04 Precise Pangolin kernels, which include the 3.2.0 and the 3.5.0 varieties.

1. sudo echo 'APT::Default-Release "trusty";' >> /etc/apt/apt.conf.d/01ubuntu

(This will make sure if the Trusty repositories have a package or update, you will get it from there.)

2. sudo echo 'deb precise-security main' >> /etc/apt/sources.list

(This adds the Precise repositories as an option for apt.)

3. Search for the latest kernel security release for the 3.2 or 3.5 line:
    #apt-get update
    #apt-cache search linux-image-3.5  -or- #apt-cache search linux-image-3.2

4. Install the desired kernel and headers:
    #apt-get install linux-image-3.5.0-54-generic linux-headers-3.5.0-54-generic

(these are just examples- use the actual version number of the kernel and header packages you want from your apt-cache search.)

5. Make sure there is not any newer kernel by apt-get remove, or specify the kernel you want to boot to in /etc/default/grub (see this article for details).

6. Reboot. Test.

7. If you're still getting the same error, it's probably not for the same reason and you didn't need to do any of this stuff. You may be dealing with another service like postfix, apache, or samba stealing your ports.  See this article about that.

Dec 14, 2014, 12:14 PM
46 Posts
Looks like a patched kernel has been created for BINDSOCK error in Ubuntu kernel 3.13
If you need it, test it and let the developer know, so we can get this into the 3.13 mainline!

Andy Whitcroft (apw) wrote on 2014-12-08:

After a long heiatus, I have a possible fix, for this which I could do some feedback on. If someone who can reproduce this could test the kernel below and firstly let me know if it fixes the issue, and secondly attach a dmesg from it. Kernels are at the URL below:

Dec 19, 2014, 1:00 PM
46 Posts
Kernel develper has isolated the bug in bindsock - PMR 41425,227,000 created
In the kernel bug (, the kernel developer who is patching the kernel for binsock issues says the root issue is in bindsock, not in the kernel:

@Ben -- the code is clear, the issue is the API is not clear. These issues have occurred because the userspace program is passing in junk in one of the fields of the structure it passes to the kernel, literally random bits from its stack. In attempting to validate those to prevent security issues this userspace application has been caught out. The main issue is the documentation for the call can be read to say you do not need to fill in that field under some circumstances, a failure in the documentation, but given that the validation needs to be more targetted; and this final fix does that, zapping the "not needed to be filled value" to zero when it is not required to avoid validation failures. The new code also documents this ABI weakness so that it should not occur.

Of course none of that excuses the userspace programmer from not initialising this structure sensibly regardless of the documentation. It is plain sloppy practice.

And also has provided a way to confirm the bug:
The attached test.c should tickle this bug, sendmsg should return ret=4 errno=0 when the fix is applied, ret=-1 errno=22 when it is not.

I just opened up PMR 41425 227 000 to try to get this fixed for good.

Jan 6, 2015, 5:39 PM
46 Posts
Update: SPR Created
"Unfortunately our level 3 support team will not pursue this issue seeing how it is not seen on current supported platforms of Unix."

So there you have it. The bindsock binary will never get fixed.

If you're having the issue, you'll need to look at one of the workarounds mentioned previously in this thread.

EDIT: From the PMR:

"A SPR (Software Problem Report) has been opened for the issue. SPR # YXYX9RA56Z "Error - Unable to Bind port 443 or 80" on SUSE12.
Jan 8, 2015, 10:08 AM
1 Posts
Simple Solution....

SLES 12 + Domino 9.0.1:

bindsock does not work only port number<1024!

1, change default ports (for example http 80->8080  443->8443 ..) 

2, Susefirewall rule: FW_REDIRECT="0/0,server ip address,tcp,80,8080"


It works for me.....



Feb 17, 2015, 11:30 AM
46 Posts
Bindsock Fixed in 9.01 FP3 IF1
SPR YXYX9RA56Z "HTTP server can't be started with "Error - Unable to Bind port 443 or 80" on SUSE12" has been resolved in 9.01 FP3 IF1 which includes a fixed bindock! Just tested on Ubuntu with 3.13 kernel, SMTP is binding and working properly there as well (unofficially, of course.) ;-)

This forum is closed to new posts and responses. New discussions are now taking place in the IBM Developer Answers forum.