| Bug #88312 | Unhelpful message: "nparts: 0 count: 1 noutside: 0" | ||
|---|---|---|---|
| Submitted: | 1 Nov 2017 8:29 | Modified: | 2 Nov 2017 6:15 |
| Reporter: | Daniël van Eeden (OCA) | Email Updates: | |
| Status: | Verified | Impact on me: | |
| Category: | MySQL Cluster: Cluster (NDB) storage engine | Severity: | S3 (Non-critical) |
| Version: | 7.6.3 | OS: | Any |
| Assigned to: | CPU Architecture: | Any | |
[1 Nov 2017 9:28]
Daniël van Eeden
Note that there were many messages, most with count: 1, but I also saw count: 2, count: 3 and count: 4.
[2 Nov 2017 6:15]
MySQL Verification Team
Hello Daniël, Thank you for the report and enhancement request! Thanks, Umesh
Description: While running ALTER TABLE t1 ALGORITHM=INPLACE, REORGANIZE PARTITION; (which failed because of 'Send Buffers overloaded in NDB kernel') I noticed these messages: "nparts: 0 count: 1 noutside: 0" These seem to be generated by NdbEventOperationImpl::readBlobParts() 561 if (unlikely(nparts != count)) 562 { 563 ndbout_c("nparts: %u count: %u noutside: %u", nparts, count, noutside); 564 } 565 assert(nparts == count); How to repeat: Check sourcecode Suggested fix: Make the message a bit easier to understand. 1. Indicate that this is from NDB 2. Indicate that this is a warning 3. Indicate that this is related to blobs 4. Indicate that this is related to a specific action/table 5. Suggest an action: Reload table, report a bug, etc.