RunPod•3d ago

CA data center connectivity issue

Hi, Having trouble with any running instance on CA data center downloading models from Cvitiai. Note this is not a authorization issue, the requests time out. The same request works perfectly fine from any other data center. Sample request wget -O epiCRealismXL.safetensors "https://civitai.com/api/download/models/128078?type=Model&format=SafeTensor&size=pruned&fp=fp16" --content-disposition

9 Replies

AstraliteHeart•3d ago

I may be having a similar issue, I am also on CA and downloading from Cloudflare R2 and all my upload/download requests are timing off right now, I can connect to the instance over the web shell and I can run same requests from my local machine with no issues. I wonder if this can be R2 related (perhaps civit is also using them under the hood) as they had an outage yesterday and something got messed up in the routing (but no idea if Cloudflare or RunPod side.

MiladOP•3d ago

Good point, this very well might be a Cloudflare issue, seems like they had another issue today with forward headers being set incorrectly

beatsbyadz•3d ago

Yea same

AstraliteHeart•3d ago

I rented another machine, I has not been able to get a CA one so far but the US one I just got is showing this

beatsbyadz•3d ago

Think its r2

AstraliteHeart•3d ago

root@79b4aca1b0a2:/workspace# traceroute google.com
traceroute to google.com (142.251.40.142), 30 hops max, 60 byte packets
 1  192.168.240.1 (192.168.240.1)  0.081 ms  0.020 ms  0.018 ms
 2  192.168.1.1 (192.168.1.1)  0.247 ms  0.179 ms  0.176 ms
 3  a-141.193.30.25.cloud.iprsecure.net (141.193.30.25)  1.525 ms  1.488 ms  1.487 ms
 4  te0-0-2-0.rcr11.ilg01.atlas.cogentco.com (38.104.64.17)  7.377 ms  7.500 ms  7.521 ms
 5  be2038.rcr22.phl01.atlas.cogentco.com (154.24.0.249)  7.297 ms be2039.rcr21.phl01.atlas.cogentco.com (154.24.5.49)  7.907 ms  7.702 ms
 6  be2364.ccr41.jfk02.atlas.cogentco.com (154.54.3.141)  9.741 ms  5.532 ms  5.232 ms
 7  be3295.ccr31.jfk05.atlas.cogentco.com (154.54.80.2)  5.841 ms be3294.ccr31.jfk05.atlas.cogentco.com (154.54.47.218)  4.631 ms  4.591 ms
 8  * 154.54.11.202 (154.54.11.202)  5.213 ms *
 9  if-be-9-2.ecore1.n75-newyork.as6453.net (63.243.128.122)  4.136 ms  4.315 ms  4.031 ms
10  72.14.221.146 (72.14.221.146)  9.912 ms  9.777 ms  9.525 ms
11  * * *
12  lga25s80-in-f14.1e100.net (142.251.40.142)  4.598 ms 142.251.60.228 (142.251.60.228)  5.090 ms  5.415 ms

root@79b4aca1b0a2:/workspace# traceroute google.com
traceroute to google.com (142.251.40.142), 30 hops max, 60 byte packets
 1  192.168.240.1 (192.168.240.1)  0.081 ms  0.020 ms  0.018 ms
 2  192.168.1.1 (192.168.1.1)  0.247 ms  0.179 ms  0.176 ms
 3  a-141.193.30.25.cloud.iprsecure.net (141.193.30.25)  1.525 ms  1.488 ms  1.487 ms
 4  te0-0-2-0.rcr11.ilg01.atlas.cogentco.com (38.104.64.17)  7.377 ms  7.500 ms  7.521 ms
 5  be2038.rcr22.phl01.atlas.cogentco.com (154.24.0.249)  7.297 ms be2039.rcr21.phl01.atlas.cogentco.com (154.24.5.49)  7.907 ms  7.702 ms
 6  be2364.ccr41.jfk02.atlas.cogentco.com (154.54.3.141)  9.741 ms  5.532 ms  5.232 ms
 7  be3295.ccr31.jfk05.atlas.cogentco.com (154.54.80.2)  5.841 ms be3294.ccr31.jfk05.atlas.cogentco.com (154.54.47.218)  4.631 ms  4.591 ms
 8  * 154.54.11.202 (154.54.11.202)  5.213 ms *
 9  if-be-9-2.ecore1.n75-newyork.as6453.net (63.243.128.122)  4.136 ms  4.315 ms  4.031 ms
10  72.14.221.146 (72.14.221.146)  9.912 ms  9.777 ms  9.525 ms
11  * * *
12  lga25s80-in-f14.1e100.net (142.251.40.142)  4.598 ms 142.251.60.228 (142.251.60.228)  5.090 ms  5.415 ms

I assume they are uplinked to "cogentco" which is telling me:

AstraliteHeart•3d ago

No idea if that covers RunPod 🙂 oh, great news, things are working for me again @Milad give it a go now

MiladOP•3d ago

Thanks, a bit of a hit and miss but it works sometimes

Gaming

Programming

CA data center connectivity issue

Did you find this page helpful?