Hello community,
I'm having a case where a user is experiencing slow uploads to GitHub, even though our upload rate is 2GB. I don't think the problem is the link or QoS policy, since we don't use it.
In order to find out the cause, I performed the following troubleshooting:
- I disabled the Threat Prevent blades: no success.
- I created a policy at the top of the rule base allowing all traffic to GitHub: no effect.
- I checked for blocks using the fw ctl zdebug + drop: no blocks.
However, when adding the traffic to fast accel, the upload rate improved significantly. Before, it took about 50 seconds to upload the file, and when adding it to fast accel, it took about 5 seconds for the same file.
Given this, I opened a case with Check Point to ask why this behavior was occurring, and Check Point responded by saying that this was normal because the traffic was undergoing several inspections, which resulted in the upload delay.
However, I have a small problem. GitHub has hundreds of IPs and I can't add them all to Fast Accel. Check Point informed me that I could add the IP of the host that needs to access GitHub, allowing everything to be accessed in Fast Accel. However, I don't see this as a solution because it will stop inspecting all the traffic from this host and may leave a gap in the network with it vulnerable.
Do you have any other ideas that I can use to solve this issue, or should I really follow the recommendations that Check Point provided?
