This message was deleted.
# general
s
This message was deleted.
b
can you file an issue for this in
pulumi/pulumi
? i'll take a look today
f
@billowy-army-68599 I think it’s b/c we don’t copy over
pulumi-analyzer-policy
in the Dockerfile
Probably need to add that along w/ the corresponding
pulumi-analyzer-policy-python
for the Python image
b
yeah I must have missed that, I'll push a fix shortly
b
thanks for the response! just created an issue @billowy-army-68599 https://github.com/pulumi/pulumi/issues/5500 let me know if there's anything I can do to help
ah i can see you've already pushed some changes that might fix this - i'll give it a try now and let you know how I go
f
Oh, sorry, I just replied on GH but now I see you updated here as well 🙂
And already found the merged PR
b
it appears to work now, however the pulumi command is now outputting a huge amount of what looks like stream debugging information during the operation, which also seems to cause it to take much longer than usual. the command completes successfully though. I ran a
pulumi preview
in the new image (sha hash
43953e33b56
) and got over 100k lines of stuff similar to the following:
Copy code
send: reset nghttp2_active_outbound_item
    send: aob->item = 0x55bae5be9b28
    stream: adjusting kept idle streams num_idle_streams=0, max=100
    send: next frame: payloadlen=8, type=7, flags=0x00, stream_id=0
    send: start transmitting frame type=7, length=17
    stream: adjusting kept idle streams num_idle_streams=0, max=100
    send: end transmission of a frame
has a debug build of the pulumi cli been used or something?
f
Not to my knowledge — that’s very odd. Sorry to replace one problem with another. We’ll have to look more into that.
b
no worries - i'll post to the github issue as well
f
Do you mind trying w/
2.11.2
which should also have the policy pack analyzer binary copied over correctly? When trying w/ that tag, I wasn’t able to repro the behavior above.
b
do you mean using the
pulumi/pulumi-nodejs:2.11.2
image?
f
sorry, yes, that’s what i meant
b
yep, that appears to work fine, so looks like the issue is specific to the
43953e33b563
image (currently tagged with
latest
and
2.11.2-ubi
)
is
latest
meant to always be the
ubi
image? previously it was the debian one iirc
f
yeah, that’s what i thought as well. I’ll have to defer to @billowy-army-68599 on that
b
no worries. thanks for your help, I understand its quite late for you guys over there 🙂 i can now work around this using the
2.11.2
tag so i'm unblocked. let me know if you need any more information from me
👍 1
f
Your realization seems to be the issue btw, 2.11.2-ubi has the same behavior
Or rather, I should say provides a hint as to where the issue might lie
b
yeah it looks like there might be some issues with how the images get tagged. i noticed at some stage yesterday that
latest
was pointing to a really old version (like
2.4.0
or something) but i think a recent update fixed that to be the actual latest version. but still looks like
latest
is now sometimes being applied to the ubi image
b
I’ll take a look at this Monday, sorry for the inconvenience
b
all good, appreciate the prompt response/fixes 👍