Michal Gow's questions

I access two Meraki VPNs from my Windows 10 Pro (10.0.19042 Build 19042) machine:

• One which is not behind a NAT - when I switch this on, I can do git clone [...] or git fetch [...] just fine.
• Second, which is behind a NAT - when I switch that on and run git fetch, I am geting error message: "fatal: unable to access 'https://bitbucket.org/[project]/[project-name].git/': gnutls_handshake() failed: Error in the pull function."

To make the second VPN work, I've executed following commands in the PowerShell:

• Set-ItemProperty -Path "HKLM:SYSTEM\CurrentControlSet\Services\PolicyAgent" -Name "AssumeUDPEncapsulationContextOnSendRule" -Type DWORD -Value 2 –Force;
• reg add "HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services\Rasman\Parameters" /v AllowL2TPWeakCrypto /t REG_DWORD /d 1 /f
• reg add "HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services\Rasman\Parameters" /v ProhibitIpSec /t REG_DWORD /d 0 /f

It was also necessary to disable IPv6.

That made the VPN work for Windows, but not for Linux inside WSL2. Any suggestions? Thank you very much!

After installation of TEX Live 2013 on (I admit outdated) openSUSE 11.3 (x86_64) I needed to add path /usr/local/texlive/2013/bin/x86_64-linux into PATH environment variables for:

• All users
• Apache

Afters hours of googling I ended up with:

• Creating file /etc/profile.d/texlive.sh with content export PATH=$PATH:/usr/local/texlive/2013/bin/x86_64-linux did the trick for the users after computer's restart.
• Adding (as a last) row export PATH=$PATH:/usr/local/texlive/2013/bin/x86_64-linux into /etc/sysconfig/apache2 worked for Apache.

It works, however I am unsure with 2 things:

1. Was it really necessary to restart server after adding file into /etc/profile.d? Isn't there some way to just reload environment?
2. It is correct to change Apache's config file when PATH environment variable is changed? Isn't there way how to simply change this variable before Apache is started?

Thank for any answers!

I have few Drupal 6 websites on webhosting, which causes this strange problem:

• some tables, especially Cache and Watchdog, tend to overhead,
• when overhead is bigger than some amount of kB, MySQL server is refusing connection to given Drupal database or connection is broken during query execution,
• Optimizing table (just overheaded rows) in phpMyAdmin is putting all back to normal.

But - until database is optimized, site is showing just MySQL errors, which is ugly... Where is a problem?

Thank you for any hints I could pass back to the hosting admins!

I recently moved a domain name in the Plesk 10 control panel from one subscription to another.

DNS is managed by another server so there is no change in DNS at all. The IP address is shared and remains the same so there is really no need for a DNS change.

But the domain is still managed by its former subscription (where it should have been removed from) and is pointing to exactly the same folder as in the past, even when is not in the list of domains there.

Subsequently, the new subscription do not have the domain under its control. Even this domain is in list of domains here and points to another folder.