user does not login within configured time(60s) login is aborted

Bug #1833469 reported by Gopinath
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
StarlingX
Fix Released
Low
zhipeng liu

Bug Description

Brief Description
-----------------
When user tries to login and the password timeout at 60s and now it takes 120s to abort.

Severity
--------
<Minor: System/Feature is usable with minor issue>

Steps to Reproduce
------------------
1.Establish a ssh connection to controller-0 terminal(CMD:ssh wrsroot@10.10.10.3)
2.It prompts for password wait for 60s and after that enter password

Expected Behavior
------------------
After 60seconds when password is entered it should abort with error
Error Message:Connection closed by 10.10.10.3 port 22

Actual Behavior
----------------
The connection is closed only after 120s.

Reproducibility
---------------
<Reproducible/Intermittent>
State if the issue is 100% reproducible or intermittent. If it is intermittent, state the frequency of occurrence

System Configuration
--------------------
< Dedicated storage>

Branch/Pull Time/Commit
-----------------------
OS="centos"
SW_VERSION="19.01"
BUILD_TARGET="Host Installer"
BUILD_TYPE="Formal"
BUILD_ID="20190613T013000Z"

JOB="STX_build_master_master"
<email address hidden>"
BUILD_NUMBER="144"
BUILD_HOST="starlingx_mirror"
BUILD_DATE="2019-06-13 01:30:00 +0000"

Last Pass
---------
Last pass information is not available now

Timestamp/Logs
--------------
[tester@registry stx-test-suite]$ ssh wrsroot@192.168.90.106
wrsroot@10.10.10.3's password:
Connection closed by 10.10.10.3 port 22

Test Activity
-------------
[ Feature Testing, Regression Testing ]

yong hu (yhu6)
Changed in starlingx:
assignee: nobody → yong hu (yhu6)
Revision history for this message
Ghada Khalil (gkhalil) wrote :

Low priority / not gating - no real system impact. ssh eventually times out as expected.

Changed in starlingx:
importance: Undecided → Low
status: New → Triaged
tags: added: stx.security
tags: added: stx.distro.other
Ada Cabrales (acabrale)
tags: added: stx.regression stx.retestneeded
Revision history for this message
Gopinath (gprabakx) wrote :
Revision history for this message
chendongqi (chen-dq) wrote :

The bug behavior is reproducible in 20190721T013000Z/outputs/iso/bootimage.iso while we connect controller0 with CMD:ssh sysadmin@10.10.10.3

the reason is that the default login timeout is 2m which is set in /etc/ssh/sshd_config
#LoginGraceTime 2m

if we want to change the default time to 1m
we should set it uncommented to 1m as below:
=> LoginGraceTime 1m

then it will take 60s to timeout, and bug behavior will not reproducible any more.

Revision history for this message
Cindy Xie (xxie1) wrote :

@chendongqi, if this is the case, please submit a patch to get the issue fixed.

Revision history for this message
Long.Li (long.li) wrote :

As we investigate, the issue is sshd_config not built in openssh-7.4p1-16.el7_4.tis.10.x86_64.rpm

when we extract openssh-7.4p1-16.el7_4.tis.10.x86_64.rpm, there is no /etc/ssh/sshd_config in it.

so in controller0, the /etc/ssh/sshd_config is the centos original file which default timeout is 2m. and not replaced by starlingx sshd_config file which is 1m.

so we suppose it is due to the openssh_config.spec not works well.

we will debug the openssh-7.4pl rpm build procedure.
then update if any progress.

yong hu (yhu6)
Changed in starlingx:
assignee: yong hu (yhu6) → nobody
assignee: nobody → Long.Li (long.li)
zhao.shuai (zhao.shuai)
Changed in starlingx:
assignee: Long.Li (long.li) → zhao.shuai (zhao.shuai)
zhipeng liu (zhipengs)
Changed in starlingx:
assignee: zhao.shuai (zhao.shuai) → zhipeng liu (zhipengs)
Changed in starlingx:
status: Triaged → In Progress
Revision history for this message
OpenStack Infra (hudson-openstack) wrote : Fix merged to integ (master)

Reviewed: https://review.opendev.org/676856
Committed: https://git.openstack.org/cgit/starlingx/integ/commit/?id=877f2b0c33b4e0547d8eed6f8bc610492f6a2d51
Submitter: Zuul
Branch: master

commit 877f2b0c33b4e0547d8eed6f8bc610492f6a2d51
Author: zhipengl <email address hidden>
Date: Fri Aug 16 23:27:37 2019 +0800

    Fix user does not login within configured time(60s) login is aborted

    Root cause is openssh-config need depend on openssh-clients and
    openssh-server RPM instead of openssh, so that when install
    openssh-config, it can overwrite ssh_config/sshd_config that installed
    by openssh-clients and openssh-server.

    Below test pass
    After deployment, the expected "LoginGraceTime 1m" can be seen in
    /etc/ssh/sshd_config, instead of default one.

    Closes-bug: 1833469

    Change-Id: Id59b728dfce39848f34279a5a2d75df417e9e100
    Signed-off-by: zhipengl <email address hidden>

Changed in starlingx:
status: In Progress → Fix Released
Revision history for this message
Ghada Khalil (gkhalil) wrote :

LP was fixed a long time ago; removing the stx.retestneeded tag

tags: removed: stx.retestneeded
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.