-
Notifications
You must be signed in to change notification settings - Fork 1.2k
Follow-up to read stream lag #1863
Comments
@lifeBCE the callback you give to |
I have updated the test code to use Worth notiing that I don't think any solution waits for the entire file to download as the test logic continues to download the large file after it reports that the event has been triggered. Also, my use case is the Here is the latest POC that shows an 8 second delay using either pullable or readable.
|
@lifeBCE I'm getting this: $ node slow-cat3.js
Instantiating IPFS node
Swarm listening on /p2p-circuit/ipfs/QmSeJDTNXyVtGCKyjZ79HGknuLaEdttm7VreA1cwefNa39
Swarm listening on /p2p-circuit/ip4/127.0.0.1/tcp/4001/ipfs/QmSeJDTNXyVtGCKyjZ79HGknuLaEdttm7VreA1cwefNa39
Swarm listening on /p2p-circuit/p2p-websocket-star/ipfs/QmSeJDTNXyVtGCKyjZ79HGknuLaEdttm7VreA1cwefNa39
Swarm listening on /ip4/127.0.0.1/tcp/4001/ipfs/QmSeJDTNXyVtGCKyjZ79HGknuLaEdttm7VreA1cwefNa39
IPFS node is ready
Stream has been created, waiting for data....
Stream is receiving data: 2 seconds $ cat node_modules/ipfs/package.json | json version
0.34.4 |
Scratch that, I AM seeing the issue when starting with a fresh repo. |
@alanshaw Are you using the same hash? Deleting local cache?
I am on 0.34.2 so I'll give 0.34.4 a try but didn't notice anything between the two that might account for the issue. |
Does this still happen with |
Closing as this issue is very stale. |
Version:
0.34.2
Platform:
Darwin MacBook-Pro.local 16.7.0 Darwin Kernel Version 16.7.0: Thu Jun 15 17:36:27 PDT 2017; root:xnu-3789.70.16~2/RELEASE_X86_64 x86_64
Subsystem:
Potential Bitswap but not sure
Type:
Bug
Severity:
Medium
Description:
This is a follow up to #1706
I am still seeing an average 8 second delay when streaming blocks before the stream will emit its
data
event only if thelength
argument is omitted or large.Steps to reproduce the error:
Note: May want to change the hash value to something in your test env.
Being a stream function, my expectation is that the size of the file or bytes being requested should not have an impact on the time to initial data event trigger. Worth mentioning that I am using
go-ipfs
as the websocket bootstrap server behind nginx.The text was updated successfully, but these errors were encountered: