This repository was archived by the owner on Sep 25, 2026. It is now read-only.
Repository navigation
This repository was archived by the owner on Sep 25, 2026. It is now read-only.
StorageException: Resumable upload is already complete. #666
Description
Following #647, we no longer sometimes see:
java.lang.NullPointerException: null
at com.google.cloud.storage.spi.v1.HttpStorageRpc.getCurrentUploadOffset(HttpStorageRpc.java:783)
at com.google.cloud.storage.BlobWriteChannel$1.run(BlobWriteChannel.java:125)
at java.base/java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:515)
at com.google.api.gax.retrying.DirectRetryingExecutor.submit(DirectRetryingExecutor.java:105)
at com.google.cloud.RetryHelper.run(RetryHelper.java:76)
at com.google.cloud.RetryHelper.runWithRetries(RetryHelper.java:50)
at com.google.cloud.storage.BlobWriteChannel.flushBuffer(BlobWriteChannel.java:70)
at com.google.cloud.BaseWriteChannel.close(BaseWriteChannel.java:151)
at com.google.cloud.storage.StorageImpl.createFrom(StorageImpl.java:265)
...snip...
This appeared in 1.113.4 but no longer appears in 1.113.6. Thanks for the fix.
Instead (in 1.113.6) we now sometimes see:
com.google.cloud.storage.StorageException: Resumable upload is already complete.
at com.google.cloud.storage.spi.v1.HttpStorageRpc.getCurrentUploadOffset(HttpStorageRpc.java:769)
at com.google.cloud.storage.BlobWriteChannel$1.run(BlobWriteChannel.java:125)
at java.base/java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:515)
at com.google.api.gax.retrying.DirectRetryingExecutor.submit(DirectRetryingExecutor.java:105)
at com.google.cloud.RetryHelper.run(RetryHelper.java:76)
at com.google.cloud.RetryHelper.runWithRetries(RetryHelper.java:50)
at com.google.cloud.storage.BlobWriteChannel.flushBuffer(BlobWriteChannel.java:70)
at com.google.cloud.BaseWriteChannel.close(BaseWriteChannel.java:151)
at com.google.cloud.storage.StorageImpl.createFrom(StorageImpl.java:265)
...snip...
What do you think? More tweaks to #604 required?
Thanks for your time.
Activity
frankyn commentedon Dec 16, 2020
Hi @n12c!
Thanks for filing the issue. Do you have an example use case that causes this issue?
n12c commentedon Dec 16, 2020
I don't have a self-contained reproducer; this seems quite intermittent. Generally we just call
com.google.cloud.storage.Storage#createFrom(com.google.cloud.storage.BlobInfo, java.io.InputStream, int, com.google.cloud.storage.Storage.BlobWriteOption...)with content in the range of 100 bytes to a few MiB in length. I added retry logic to my own code, which repeats this call unmodified and without delay (with a freshjava.io.InputStreamhaving the exact same content); this always seems to succeed on the second try. We always useBlobWriteOption#disableGzipContentand sometimes useBlobWriteOption#encryptionKey(java.security.Key). The only instances I have seen have all occurred while CSEK was in use, but that doesn't necessarily prove it doesn't also happen without CSEK.I can try to write a reproducer but it would basically consist of uploading random objects using this API in a tight loop until (hopefully) encountering the same issue.
frankyn commentedon Dec 16, 2020
Thanks! I'll try to reproduce the issue, it might be a retry an accidental retry when there shouldn't be one.
n12c commentedon Dec 16, 2020
Update: I just saw this happen without CSEK (no encryption of any kind, just the default Google managed keys). So, CSEK is not a factor.
richiesgr commentedon Jan 3, 2021
Hi
Any update regarding what cause this ?
frankyn commentedon Jan 11, 2021
Hi @richiesgr I'm working to release a fix today. Apologies for the delay.
fix: last chunk is retriable (#677)