View Issue Details

IDProjectCategoryView StatusLast Update
0016797CentOS-8kernelpublic2019-12-04 04:03
Reporteragabeckov 
PrioritynormalSeveritymajorReproducibilityunable to reproduce
Status newResolutionopen 
Product Version8.0.1905 
Target VersionFixed in Version 
Summary0016797: XFS runs out of inodes very quickly
DescriptionSo, for some reason root partition with XFS was created with only ~65000 inodes, and it run out of it pretty quickly. Then running services just fail to start.
Output of df -i
df -i
Filesystem Inodes IUsed IFree IUse% Mounted on
devtmpfs 991911 354 991557 1% /dev
tmpfs 1000177 1 1000176 1% /dev/shm
tmpfs 1000177 578 999599 1% /run
tmpfs 1000177 17 1000160 1% /sys/fs/cgroup
/dev/mapper/cl_zabbix-root 65576 63532 2044 97% /
/dev/mapper/cl_zabbix-home 9115648 23 9115625 1% /home
/dev/sda1 65536 316 65220 1% /boot
tmpfs 1000177 5 1000172 1% /run/user/0
Have no idea how it started - it's just happened today.
Kernel
4.18.0-80.el8.x86_64 #1 SMP Tue Jun 4 09:19:46 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux
After reboot sometimes inodes on / partition grows to ~96000, then it goes down quickly to ~65000 again. With catastrophic consequences for running applications. The host runs zabbix, mysql and apache.
xfs_growfs -m does not help.
Tagsinodes, xfs

Activities

agabeckov

agabeckov

2019-12-04 04:03

reporter   ~0035793

For some reason, once I added space to / filesystem and did xfs_growfs, it worked. But still, that was a nasty surprise.

Issue History

Date Modified Username Field Change
2019-12-04 03:27 agabeckov New Issue
2019-12-04 03:27 agabeckov Tag Attached: xfs
2019-12-04 03:27 agabeckov Tag Attached: inodes
2019-12-04 04:03 agabeckov Note Added: 0035793