'fsck -n' will show inconsistencies with EAU maps. For example
pass4 - checking resource maps
fileset 999 au 40 imap incorrect - fix (ynq)n
fileset 999 au 40 iemap incorrect - fix (ynq)n
fileset 999 au 80 imap incorrect - fix (ynq)n
fileset 999 au 103 imap incorrect - fix (ynq)n
au 45 emap incorrect - fix? (ynq)n au 45 summary incorrect - fix? (ynq)n
au 53 emap incorrect - fix? (ynq)n au 53 summary incorrect - fix? (ynq)n
The different flushing of different maps causes the on-disk filesystem to become inconsistent, if the log replay is interrupted.
The fsck log replay is updated to synchronously write secondary map and EAU summary to the disk.
The following patches have been released to fix the issue
Public Patches for Infoscale 7.3.1
| OS Version | Patch ID | Sort Link |
|---|---|---|
| SLES 12 |
Infoscale 7.3.1.100 (with VxFS 7.3.1.2600) |
https://sort.veritas.com/patch/detail/15006 |
Public Patches for Infoscale 7.4.1
| OS Version | Patch ID | Sort Link |
|---|---|---|
| RHEL 6 |
Infoscale 7.4.1.1200 (with VxFS 7.4.1.1200) |
https://sort.veritas.com/patch/detail/15048 |
| RHEL 7 |
Infoscale 7.4.1.1200 (with VxFS 7.4.1.1200) |
https://sort.veritas.com/patch/detail/15049 |
| RHEL 7.7 |
Infoscale 7.4.1.1300 (with VxFS 7.4.1.1300) |
https://sort.veritas.com/patch/detail/15132 |
Private HotFixes (contact support for private HotFixes)
| Affected VxFS version | Private HotFix |
|---|---|
| 7.2 | 7.2.0.4201 |
| 7.3.1 | 7.3.1.2601 |
Appliance EEB's
There is a separate technote 100048518 for appliances.
The issue is fixed with NBA 3.2MR2 patch.
The following EEB's are available externally from the download center via the appliance technote.
| NBA version | EEB |
|---|---|
| 3.1.0 | ET4004878 |
| 3.1.1 | ET4006111 |
| 3.1.2 | ET4012237 |
| 3.2.0 | ET4004883 |