Search results

@petethompson, have you setup the certificates at the listener level? VHost Template level is not enough.

@bicibikes15, looks like the problem is the mysql database is not up or not accessible. Best would be to reach out to Google Cloud support for help.

@jack lee, honestly this kind of "benchmarking" does not demonstrate very much for the server. They depend mostly on the website content itself (static vs dynamic), the machine it's served from, and the client making the requests.

Hi @jack lee, I'm sorry you are experiencing so much trouble setting up the reverse proxy. I am a bit confused by your last post. http://127.0.0.1:12000 should be the address/port of your existing server, which will be behind the OLS proxy. So it has to be running first. The fact that you are...

@Cold-Egg: Any chance support for Perl extended regular expressions could be added for contexts?

@Cold-Egg: It would be great if there was an example to show how to allow specific file extensions and deny all others. Using the negative match operator (?!) from Perl extended regular expressions does not seem to work.

UPDATED You can create a static context at the virtual host level using the following expression: then set Location as: and set Accessible to 'No'. Leaving the location empty should work too.

Rewrite rules won't work as OLS does not support them for file access. What file extensions are you trying to allow/deny access to? Do you want to allow access to all jpg|jpeg|png|gif|mp3|ogg files and deny everything else?

This looks like your node app is trying to read from stdin, which may not be available through LSNODE.

I have observed that OLS webadmin GUI does not ask for a restart when context is deleted. I think it extends to anything being deleted. Whereas it always asks for a restart when something is being added or changed. Feels like a minor bug.

Mentioning that you will usually want to make sure that 'Chained Certificate' is set to 'Yes'.

Hi @kowalski215, it depends where the problem is coming from. First, if you are getting good results with caching, you can use the crawler to pre-cache pages. Your visitors will see cached pages most of the time. After that, you will want to figure out whether it's OLS or a backend issue...

Thanks @Cold-Egg for this precious information. I totally missed that part of your earlier post. Wouldn't 'try-restart' (i.e. restart if already running) be better than 'restart'?

1) Have you looked at the server logs under /usr/local/lsws/logs? 2) Redis server not working is not a good sign. Have you looked at the redis server log or error messages under /var/log/syslog? 3) Are you using local DigitalOcean networking to access the database or the machine public IP? 4) I...

It's not possible at the moment. Cloudflare has disabled http/3 for the time being in some areas. Mine and probably yours too. At least, that's my understanding. See here: https://community.cloudflare.com/t/community-tip-http-3-with-quic/117551

If this comes down to a permission issue, it would be great to have some kind of error message in the server error log (if it's not there) and a permission error message in the webadmin GUI.

zip/bzip2 are included in the base package too (as far as I can tell).

Yes, looks right.

What the rule is telling you is, if you need it, add the format to this file: /usr/local/lsws/conf/modsec/comodo/userdata_wl_content_type Mine looks like this: As I have application/x-www-form-urlencoded listed, I don't think the rule would trigger in my case.

Oh I have it reversed. The POST request method is not allowed, that's why it triggered. You may have to disable the rule for this website.