[LU-4324] Write data failed , print error as "No space left on the device"

• Status: Resolved
• Project: Lustre
• Affects Version/s: Lustre 2.4.0
• Type: Bug
• Priority: Major
• Reporter: yueyuling
• Assignee: WC Triage
• Resolution: Cannot Reproduce

Environment

2 Lustre Servers + 1 Client Server

• Epic/Theme: test
• Severity: 3

Description

1.mount 1 MDT and 4 OSTs on the Lustre Server1.
2.mount 4 OSTs on the Lustre Server2.
3.Config Lustre failover between the 2 Lustre Servers.
4.mount the Lustre File System on the Lustre Client.
5.Write and Read datas on the Lustre Client.
6.At the same time, test the Failover function between the 2 Lustre Servers in turn.
7.Write data failed when the Lustre Server2 take the task of the Lustre Server1,and print error info as: "No space left on the device", but there is enouth space left on the device actually.

Comments

[Oleg Drokin]

Can you please post some logs from the systems after the error happened that shows the errors?

[yueyuling]

OK.
The logs of the problem are shown in the attachment. In addition, I think I should explain some of the details of the problem.
1.The time of the Client is 8 hours before the Lustre Servers.
2.The error info only print during the the Lustre Server2 take the task of the Lustre Server1. Other times, writing and reading datas from the Lustre FS is normal.
3.During my test, the problem appeared 7 times, the time of the Lustre Server as follows:
(1)9-17： 19:47
(2)9-17： 20:45
(3)9-17: 22:45
(4)9-18： 01:45
(5)9-18： 03:45
(6)9-18： 05:45
(7)9-18： 07:45

[Andreas Dilger]

Close old bug