Fix issues in ElasticSearch 7.14 and add log cases to ES7.14

[kezhenxu94]

• If this pull request closes/resolves/fixes an existing issue, replace the issue number. Closes Exception mapper_parsing_exception in ElasticSearch 7.14 #7425.
• Update the CHANGES log.

[wu-sheng]

Let's limit the not null in the certain scopes.

[kezhenxu94]

Let's limit the not null in the certain scopes.

The nullability check is added according to whether they are annotated with org.apache.skywalking.oap.server.core.storage.annotation.Column#matchQuery that will add a copy_to field, which causes the problem. If we skip some of the cases mentioned in the comments, the root cause (serviceName being null) is hidden behind the ElasticSearch level message, although serviceName is expected to be non null,

[codecov]

Codecov Report

Merging #7429 (c60bb4a) into master (cc66254) will decrease coverage by 0.84%.
The diff coverage is 25.00%.

❗ Current head c60bb4a differs from pull request most recent head 87b3e42. Consider uploading reports for the commit 87b3e42 to get more accurate results

@@             Coverage Diff              @@
##             master    #7429      +/-   ##
============================================
- Coverage     53.14%   52.30%   -0.85%     
+ Complexity     4335     4005     -330     
============================================
  Files          1899     1034     -865     
  Lines         41163    26449   -14714     
  Branches       4620     2620    -2000     
============================================
- Hits          21877    13833    -8044     
+ Misses        18152    11408    -6744     
- Partials       1134     1208      +74     

Impacted Files	Coverage Δ
...er/core/analysis/manual/log/AbstractLogRecord.java	45.94% <0.00%> (-4.06%)	⬇️
...browser/manual/errorlog/BrowserErrorLogRecord.java	45.16% <50.00%> (-1.51%)	⬇️
...ing/oap/server/core/storage/IHistoryDeleteDAO.java	0.00% <0.00%> (-100.00%)	⬇️
...g/oap/server/core/cluster/ClusterHealthStatus.java	0.00% <0.00%> (-100.00%)	⬇️
...p/server/core/analysis/metrics/MinLongMetrics.java	0.00% <0.00%> (-100.00%)	⬇️
...er/exporter/provider/grpc/GRPCExporterSetting.java	0.00% <0.00%> (-100.00%)	⬇️
...log/analyzer/provider/log/ILogAnalyzerService.java	0.00% <0.00%> (-100.00%)	⬇️
...king/oap/server/configuration/api/ConfigTable.java	0.00% <0.00%> (-100.00%)	⬇️
...analyzer/provider/meter/process/SampleBuilder.java	0.00% <0.00%> (-100.00%)	⬇️
...skywalking/oap/server/receiver/envoy/als/Role.java	0.00% <0.00%> (-100.00%)	⬇️
... and 1378 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update cc66254...87b3e42. Read the comment docs.

[wu-sheng]

Let's limit the not null in the certain scopes.

The nullability check is added according to whether they are annotated with org.apache.skywalking.oap.server.core.storage.annotation.Column#matchQuery that will add a copy_to field, which causes the problem. If we skip some of the cases mentioned in the comments, the root cause (serviceName being null) is hidden behind the ElasticSearch level message, although serviceName is expected to be non null,

The changes on the log side are reasonable, and should be done. But why we bother to check these should-not-null fields? If you want to check and expose the cause, check on the source#prepare stage rather than here.