Replication fail due to invalid_json error

[huneau]

Description

I'm trying to move my couchdb to another location.
I dont use partition.
And I've got that :

[error] 2024-03-14T11:07:53.352189Z nonode@nohost <0.28648.0> -------- gen_server <0.28648.0> terminated with reason: bad return value {invalid_json,{1,invalid_json}}
2024-03-14T11:07:53.352507429Z   last msg: redacted
2024-03-14T11:07:53.352512397Z      state: [{main_pid,<0.28613.0>},{loop,<0.28649.0>},{source,"http:https://xxxx/db/"},{target,"https://xxx/db/"},{num_readers,20},{pending_fetch,{{<0.28649.0>,#Ref<0.62576232.3310616584.252018>},{<<"0af1d675d50e6fadaaf1db7d25307baa">>,[{1,<<239,151,123,103,148,227,196,120,165,221,156,249,165,52,34,38>>}],[]}}},{batch_size,79789}]
2024-03-14T11:07:53.352523876Z     extra: [<0.31236.0>]
2024-03-14T11:07:53.352531059Z [error] 2024-03-14T11:07:53.352267Z nonode@nohost <0.28648.0> -------- gen_server <0.28648.0> terminated with reason: bad return value {invalid_json,{1,invalid_json}}
2024-03-14T11:07:53.352534232Z   last msg: redacted
2024-03-14T11:07:53.352536389Z      state: [{main_pid,<0.28613.0>},{loop,<0.28649.0>},{source,"http:https://xxxx/db/"},{target,"https://xxxxx/db/"},{num_readers,20},{pending_fetch,{{<0.28649.0>,#Ref<0.62576232.3310616584.252018>},{<<"0af1d675d50e6fadaaf1db7d25307baa">>,[{1,<<239,151,123,103,148,227,196,120,165,221,156,249,165,52,34,38>>}],[]}}},{batch_size,79789}]
2024-03-14T11:07:53.352539970Z     extra: [<0.31236.0>]
2024-03-14T11:07:53.363451648Z [error] 2024-03-14T11:07:53.363238Z nonode@nohost <0.28613.0> -------- Worker <0.28648.0> died with reason: {bad_return_value,{invalid_json,{1,invalid_json}}}
2024-03-14T11:07:53.363984631Z [error] 2024-03-14T11:07:53.363328Z nonode@nohost <0.28613.0> -------- Replication `00edc48d343c8376818d7bb84b49566a+continuous` (`http:https://xxxx/db/` -> `https://xxxx/db/`) failed: {worker_died,<0.28648.0>,{bad_return_value,{invalid_json,{1,invalid_json}}}}

Is there a way to debug that ?

Your Environment

Source : couchdb 3.2.2 on Docker
Dest : couchdb 3.3.3 on Docker

[huneau]

Finally found my issue, we have in our DB some massive Documents with conflicts that cause issue with network.
That's it, hopefully, the migration will end up with success.

[nickva]

No worries at all. You can try to reduce the replicator batch size: https://docs.couchdb.org/en/stable/config/replicator.html#replicator/worker_batch_size if you have huge documents. Make it 50 instead of the default 500, for instance. If you see timeouts, consider increasing https://docs.couchdb.org/en/stable/config/replicator.html#replicator/connection_timeout