這個主題有2篇回覆

[Chase]

For some reason today, new notes started having issues when added via the Python SDK. Here's where the error is being raised. The function that's being called is createNote().

Relevant portion of trace:


createdNote = noteStore.createNote(authToken,note)
File "/usr/local/lib/python2.6/dist-packages/evernote/edam/notestore/NoteStore.py", line 4695, in createNote
return self.recv_createNote()
File "/usr/local/lib/python2.6/dist-packages/evernote/edam/notestore/NoteStore.py", line 4707, in recv_createNote
(fname, mtype, rseqid) = self._iprot.readMessageBegin()
File "/usr/local/lib/python2.6/dist-packages/thrift/protocol/TBinaryProtocol.py", line 126, in readMessageBegin
sz = self.readI32()
File "/usr/local/lib/python2.6/dist-packages/thrift/protocol/TBinaryProtocol.py", line 203, in readI32
buff = self.trans.readAll(4)
File "/usr/local/lib/python2.6/dist-packages/thrift/transport/TTransport.py", line 63, in readAll
raise EOFError()
EOFError

Any ideas as to why this might be happening?

[SethH]

I don't think that this is a generic problem with the Python SDK - I'm able to run our sample code without a problem. Does the EDAMTest.py sample code work for you? What platform are you on?

[Chase]

Hi Seth! The weird thing is this was happening in a 2 hour period earlier when we had really high disk I/O. It's not happening anymore now that I look at the logs closely. This is running on Ubuntu 10.04 LTS in a VPS. The code is battle-tested and just started doing this randomly. Definitely not a problem with the SDK in general I'd imagine, was just wondering if you'd have ideas. It seems like it's tied to our high disk I/O since they were happening around the same time. Not sure why that would be though.