Anyone that has experience `pulumi up` failing by ...
# general
w
Anyone that has experience
pulumi up
failing by being "killed", it happens my all the time now in one of our projects:
Copy code
➜  deploy git:(deps/minimist_1.2.7) ✗ pulumi up
View Live: https://.....

[1]    3433262 killed     pulumi up
➜  deploy git:(deps/minimist_1.2.7) ✗
this happens both on the local machine and on my build server.
To get it working again I had to remove the stack all together and then create it again, not ideal but worked this time. I tried to do a verbose debug during
pulumi up
but there was no relevant information as I could see.
e
Not seen this, if it repros you might want to run with debug logs on
--logtostderr -v10
to see what prints from that
w
I did v9, but nothing useful.
e
Might be out of memory? Is this a large stack, or using a lot of nested parent resources?
w
The stack was pretty small, so I’m pretty sure that isn’t the case.
e
Might be worth checking that, try running top while running up and keeping an eye on usage, the only time I've seen pulumi getting killed like this has been the oom killer
w
As mentioned, I was washer trying to fix it yesterday and I did by doing a rm on the stack and then try again which lead me to believe that there was something related to the pulumi service.
e
oh interesting, I'll keep an eye out for anything like this coming up again. There was some service latency the other day but I wouldn't expect it to trigger process kills
w
Don't think it was a service latency either since other updates worked fine, it was just this one that failed over and over again.