Bugzilla will be upgraded to version 5.0. The upgrade date is tentatively scheduled for 2 December 2018, pending final testing and feedback.
First Last Prev Next    This bug is not in your last search results.
Bug 1408572 - tmpfs not mounted to /tmp
tmpfs not mounted to /tmp
Status: CLOSED ERRATA
Product: Red Hat Enterprise Linux 7
Classification: Red Hat
Component: releng (Show other bugs)
7.3
Unspecified Unspecified
unspecified Severity unspecified
: rc
: ---
Assigned To: Lubos Kocman
Release Test Team
:
Depends On:
Blocks:
  Show dependency tree / graph
 
Reported: 2016-12-25 02:25 EST by Sergey Safarov
Modified: 2017-08-01 13:32 EDT (History)
6 users (show)

See Also:
Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of:
Environment:
Last Closed: 2017-08-01 13:32:24 EDT
Type: Bug
Regression: ---
Mount Type: ---
Documentation: ---
CRM:
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---

Attachments (Terms of Use)
tmp mount info (1.39 KB, text/plain)
2017-01-02 08:47 EST, Sergey Safarov 		no flags Details
tmp mount info 2 (2.62 KB, text/plain)
2017-01-02 13:00 EST, Sergey Safarov 		no flags Details
boot log (1.40 MB, text/plain)
2017-01-04 09:27 EST, Sergey Safarov 		no flags Details
Proposed patch (1.35 KB, patch)
2017-01-05 10:51 EST, Michal Sekletar 		no flags Details | Diff
Add an attachment (proposed patch, testcase, etc.)


External Trackers
Tracker ID Priority Status Summary Last Updated
Red Hat Product Errata RHBA-2017:1850 normal SHIPPED_LIVE redhat-release update 2017-08-01 14:03:01 EDT

  None (edit)
Description Sergey Safarov 2016-12-25 02:25:41 EST 
Description of problem:
I created fstab file

[root@ip-172-31-39-209 ~]# cat /etc/fstab 
UUID=ef6ba050-6cdc-416a-9380-c14304d0d206 /                       xfs     defaults        0 0
tmpfs	/tmp    tmpfs   size=256m    0 0

After reboot tmpfs not mounted to /tmp path.
If i give command "mount /tmp" then executed correctly

If i change "/tmp" to "/tmp2" and will create "/tmp2 folder, then all will be correct.


Expected results:

After reboot correctly mounted /tmp volume.
Comment 2 Lukáš Nykrýn 2017-01-02 07:31:43 EST 
What is your version of systemd?

Works just fine on my machine:

[0 root@qeos-225 ~]# mount | grep '/tmp'
tmpfs on /tmp type tmpfs (rw,relatime,seclabel,size=262144k)
[0 root@qeos-225 ~]# grep '/tmp' /etc/fstab
tmpfs	/tmp    tmpfs   size=256m    0 0
[0 root@qeos-225 ~]# rpm -q systemd
systemd-219-30.el7.x86_64

Also could you attach output of "systemctl show tmp.mount"?
Comment 3 Sergey Safarov 2017-01-02 08:46:50 EST 
Requested info in attached file.
Comment 4 Sergey Safarov 2017-01-02 08:47 EST 
Created attachment 1236663 [details]
tmp mount info
Comment 5 Lukáš Nykrýn 2017-01-02 09:16:51 EST 
hmmm weird.
Can you check that /run/systemd/generator/tmp.mount and /run/systemd/generator/local-fs.target.requires/tmp.mount exist?
What about "systemctl show local-fs.target"?
Comment 6 Sergey Safarov 2017-01-02 13:00 EST 
Created attachment 1236701 [details]
tmp mount info 2

Requested info is attached in file tmp-mount-info2.txt
Comment 7 Lukáš Nykrýn 2017-01-03 03:41:14 EST 
Hmm that seems to be fine. Can you try to reboot your machine, put the word "debug" on kernel cmdline and post here the output of "journalctl -b" after the machine boots up?
Comment 8 Sergey Safarov 2017-01-04 09:27 EST 
Created attachment 1237190 [details]
boot log

Requested boot log in attached file "boot-debug.txt"
Comment 9 Michal Sekletar 2017-01-04 12:37:00 EST 
I tried this with unmodified CentOS7 vagrant box (except updated systemd) and I couldn't reproduce. From logs I can see that you are running the VM machine in EC2. I'd bet that there is something fishy with how that centos7 AMI is built. I'll be able to test this on EC2 tomorrow.
Comment 10 Michal Sekletar 2017-01-05 07:47:33 EST 
I was able to reproduce this problem on EC2 with both RHEL7 and CentOS7. In both images tmp.mount is masked by default, hence mount unit generated from fstab doesn't take any effect. 

This is a configuration issue and not a bug in systemd. We should talk to people building those AMIs and advice them to stop masking tmp.mount. In the meantime please unmask the unit manually in your provisioning scripts.
Comment 11 Michal Sekletar 2017-01-05 10:51 EST 
Created attachment 1237723 [details]
Proposed patch
Comment 12 Lubos Kocman 2017-01-24 08:04:28 EST 
Modified in spin-kickstarts

dhcp-24-203:rhel7 lkocman$ git status
On branch master
Your branch is up-to-date with 'origin/master'.
Changes to be committed:
  (use "git reset HEAD <file>..." to unstage)

	modified:   rhel-7.3-server-ec2.ks
	modified:   rhel-7.3-server-kvm.ks
	new file:   rhel-7.4-server-ec2.ks
	modified:   rhel-7.4-server-kvm.ks
	modified:   rhel-7.4-server-rt.ks

Change should be visible on next spin of image.

Lubos
Comment 15 errata-xmlrpc 2017-08-01 13:32:24 EDT 
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2017:1850
Note You need to log in before you can comment on or make changes to this bug.
First Last Prev Next    This bug is not in your last search results.