我需要修复一些BQ表，唯一的方法似乎是导出它们，修复本地和重新上载。但我所有的出口尝试都失败了。以下只是一个例子。我正在使用命令行bq。
bq extract --compression=GZIP --destination_format=NEWLINE_DELIMITED_JSON test.US_2015_05 gs://???/us-2015-05--*.gz
BigQuery error in extract operation: Error processing job '???:bqjob_r1dbda9e5_00000150b2e288eb_1': An internal error occurred and the request could not be completed.
如果我从URI中删除了通配符，我会收到不同的错误。
'???:bqjob_r3e0f4cf0_00000150b2e6daf6_1': Table gs://???/us-2015-05.json too large to be exported to a single file. Specify a uri including a * to shard export.
此外，我无法使用默认的CSV导出，因为我的表格中包含REPEATED字段。所以它必须是JSON（压缩或不压缩）。
有什么想法吗？
[编辑]如果我尝试从网络用户界面导出，我会收到相同的内部错误。
答案 0 :(得分：1)
After a lot of fiddling back and forth the solution is incredibly simple and yet totally unaligned with the vanilla "internal error".
The bucket I tried to export to was in EU region and this seems to confuse BQ. Creating a brand new bucket in US region made it work like a charm.
Now, how's this an "internal error" is beyond me. Nevertheless, I am glad I got this to work. I hope this helps someone save up a few wasted hours.
答案 1 :(得分：1)
＆＃34;内部错误＆＃34;是我们的跨区域执法中的一个错误：我们的错误地图中缺少一个欧洲地区，您碰巧偶然发现了它。很抱歉给您带来不便，我们很快就会解决此问题。