Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.
RHEL Engineering is moving the tracking of its product development work on RHEL 6 through RHEL 9 to Red Hat Jira (issues.redhat.com). If you're a Red Hat customer, please continue to file support cases via the Red Hat customer portal. If you're not, please head to the "RHEL project" in Red Hat Jira and file new tickets here. Individual Bugzilla bugs in the statuses "NEW", "ASSIGNED", and "POST" are being migrated throughout September 2023. Bugs of Red Hat partners with an assigned Engineering Partner Manager (EPM) are migrated in late September as per pre-agreed dates. Bugs against components "kernel", "kernel-rt", and "kpatch" are only migrated if still in "NEW" or "ASSIGNED". If you cannot log in to RH Jira, please consult article #7032570. That failing, please send an e-mail to the RH Jira admins at rh-issues@redhat.com to troubleshoot your issue as a user management inquiry. The email creates a ServiceNow ticket with Red Hat. Individual Bugzilla bugs that are migrated will be moved to status "CLOSED", resolution "MIGRATED", and set with "MigratedToJIRA" in "Keywords". The link to the successor Jira issue will be found under "Links", have a little "two-footprint" icon next to it, and direct you to the "RHEL project" in Red Hat Jira (issue links are of type "https://issues.redhat.com/browse/RHEL-XXXX", where "X" is a digit). This same link will be available in a blue banner at the top of the page informing you that that bug has been migrated.

Bug 1523240

Summary: Overflow leading to corruption after 160E writes
Product: Red Hat Enterprise Linux 7 Reporter: Sweet Tea Dorminy <sweettea>
Component: kmod-kvdoAssignee: Sweet Tea Dorminy <sweettea>
Status: CLOSED ERRATA QA Contact: Jakub Krysl <jkrysl>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 7.5CC: awalsh, jkrysl, limershe, lmiksik
Target Milestone: rc   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: 6.1.0.99 Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2018-04-10 16:26:33 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Sweet Tea Dorminy 2017-12-07 13:43:04 UTC
Description of problem:

After 160 exabytes of data written to a VDO, the journal sequence number overflows one of the locations it is stored on disk, which is 48 bits. When this overflow happens, crash recovery has a strong possibility to corrupt data. This takes 4.5 years on theoretically ideal NVMe drives with an ideal write pattern, or on the Intel Optane 900p drive about 16 years with the same pattern.

(Math:      
       On theoretical NVMe drives, with latency of 2.8 microseconds, VDO                                                                                                                                           
       could (if not CPU bound, and having a fully allocated block map tree                                                                                                                                        
       in memory, and not memory latency bound) write 8M of zeroes in                                                                                                                                              
       5.6 microseconds, or 1.4 T/s. On such hardware, we could zero a                                                                                                                                             
       max-size 4P VDO in an hour. We could zero it 40 000 times in                                                                                                                                                
       approximately 4.5 years.                                                                                                                                                                                    
                                                                                                                                                                                                                   
       4P = (1<<40) blocks. 40000 is 40<<10. Each write makes two journal                                                                                                                                          
       entries. Therefore, this is 80<<50 journal entries; at 312 (~320) entries                                                                                                                                   
        per journal block, this is 1<<48 journal blocks. 
)

Version-Release number of selected component (if applicable):

kmod-kvdo-6.1.0.81

How reproducible:

Always, given time.

Steps to Reproduce:
1. Write 160 exabytes.
2. Crash.
3. Validate your data.

Actual results:

Corruption.

Expected results:

Advance notice before crash that an upgrade to a version with a fix for this is necessary.

Additional info:

Comment 5 Andy Walsh 2017-12-13 03:11:18 UTC
This bug fix was mistakenly not merged into the version specified.  It is now pending in a new version (6.1.0.99)

Comment 7 Jakub Krysl 2018-01-02 09:10:12 UTC
As I do not have enough time to hit this, testing SanityOnly. My tests could not find any regressions, setting to verified.

Comment 10 errata-xmlrpc 2018-04-10 16:26:33 UTC
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/RHEA-2018:0900