a lots of events coming from mysql

[kwins]

大量的数据库RowsEvent，在处理过程中。我重启了一次，然后就会出现下面的错误：
Err: table id 128: invalid table id, no corresponding table map event
但是当数据量小的时候，重启则不会造成这个问题。
请问这个问题 是因为数据解析问题吗？还是其他原因？

[marlonfan]

我也出现了 Err: table id 128: invalid table id, no corresponding table map event 但是没找到原因

[guoyuanchao1202]

有什么好的解决方法吗？我也遇到了

[lance6716]

please provide more information so we can locate it.

[guoyuanchao1202]

In filepos mode, sometimes there will be errors like:
binlogstreamer.go:77 close sync with err: Header &replication.EventHeader{Timestamp:0x615714e0, EventType:0x1e, ServerID:0x2532, EventSize:0x134e, LogPos:0x122b2ad3, Flags:0x0}, Data ... Err: table id 125679: invalid table id, no corresponding table map event

we use this in our live env, so can you smart guys can help me? thx so much

[lance6716]

seems you start replication on a binlog position that is halfway inside a transaction. that is to say, maybe you don't start replication from the end of a rotate/XID event. BlueMind ***@***.***> 于 2021年10月7日周四 21:37写道：

…

In filepos mode, sometimes there will be errors like: binlogstreamer.go:77 close sync with err: Header &replication.EventHeader{Timestamp:0x615714e0, EventType:0x1e, ServerID:0x2532, EventSize:0x134e, LogPos:0x122b2ad3, Flags:0x0}, Data ... Err: table id 125679: invalid table id, no corresponding table map event we use this in our live env, so can you smart guys can help me? thx so much — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#323 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAM4RJXRMDE5KG4SQZK4XODUFWPADANCNFSM4FXQJL3A> .

[guoyuanchao1202]

seems you start replication on a binlog position that is halfway inside a transaction. that is to say, maybe you don't start replication from the end of a rotate/XID event. BlueMind @.***> 于 2021年10月7日周四 21:37写道：
…
In filepos mode, sometimes there will be errors like: binlogstreamer.go:77 close sync with err: Header &replication.EventHeader{Timestamp:0x615714e0, EventType:0x1e, ServerID:0x2532, EventSize:0x134e, LogPos:0x122b2ad3, Flags:0x0}, Data ... Err: table id 125679: invalid table id, no corresponding table map event we use this in our live env, so can you smart guys can help me? thx so much — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#323 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAM4RJXRMDE5KG4SQZK4XODUFWPADANCNFSM4FXQJL3A .

#396
Aha, does it have anything to do with this issue? Or my fault

[guoyuanchao1202]

seems you start replication on a binlog position that is halfway inside a transaction. that is to say, maybe you don't start replication from the end of a rotate/XID event. BlueMind @.***> 于 2021年10月7日周四 21:37写道：
…
In filepos mode, sometimes there will be errors like: binlogstreamer.go:77 close sync with err: Header &replication.EventHeader{Timestamp:0x615714e0, EventType:0x1e, ServerID:0x2532, EventSize:0x134e, LogPos:0x122b2ad3, Flags:0x0}, Data ... Err: table id 125679: invalid table id, no corresponding table map event we use this in our live env, so can you smart guys can help me? thx so much — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#323 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAM4RJXRMDE5KG4SQZK4XODUFWPADANCNFSM4FXQJL3A .

Because I saw someone else raised this question like me.

[lance6716]

I'm not familiar with the canal component in this repo, but I think our project using this repo works well because we restart replication from the starting position of a whole transaction when we need re-sync. if we restart halfway inside a transaction, we'll see an orphanic rows event without table_map event.

[guoyuanchao1202]

seems you start replication on a binlog position that is halfway inside a transaction. that is to say, maybe you don't start replication from the end of a rotate/XID event. BlueMind @.***> 于 2021年10月7日周四 21:37写道：
…
In filepos mode, sometimes there will be errors like: binlogstreamer.go:77 close sync with err: Header &replication.EventHeader{Timestamp:0x615714e0, EventType:0x1e, ServerID:0x2532, EventSize:0x134e, LogPos:0x122b2ad3, Flags:0x0}, Data ... Err: table id 125679: invalid table id, no corresponding table map event we use this in our live env, so can you smart guys can help me? thx so much — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#323 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAM4RJXRMDE5KG4SQZK4XODUFWPADANCNFSM4FXQJL3A .

And I use github.com/siddontang/go-mysql v1.1.0, maybe this reson?

[guoyuanchao1202]

I'm not familiar with the canal component in this repo, but I think our project using this repo works well because we restart replication from the starting position of a whole transaction when we need re-sync. if we restart halfway inside a transaction, we'll see an orphanic rows event without table_map event.

okay got it, tkx and I just found that my project use github.com/siddontang/go-mysql v1.1.0, Is it possible that the problem in the old version has not been fixed?