automaciej's questions

Optional background (to avoid the XY problem): I'm working on a static site which will contain URLs like www.example.com/książka/. The problem with those URLs is that when you share them / paste to an IM program, etc, the diacritic characters are expanded into %<hex> and become unreadable: www.example.com/ksi%C4%85%C5%BCka/. So I'm considering the following:

• setting the canonical URL of the page as www.example.com/ksiazka/ ("ąż" replaced with "az")
• returning a temporary redirect (HTTP 302 or 307) to www.example.com/książka/.

What won't work:

Automatic address rewriting. Reason: We're not going from "ąż" to "az", but from "az" to "ąż" or maybe "ąz", or "aż", or "áz", or one of many other possibilities. The redirects must be generated during page generation, when we know the complete list of existing pages, and so we know that "ksiazka" should redirect to "książka" and not "książką".

I need to configure 1 HTTP redirect per page. I will need one new HTTP redirect every time I upload a new article.

Editing the NGINX config requires root access and I'm deploying the site under a regular user's permissions. Under Apache you could add some directives in .htaccess, but this is problematic for performance, and a no-go for NGINX.

Bar that, I could create an include in /etc/nginx/sites-available/mysite to pull in contents of /home/myuser/mysite/_redirects. So the server would read in a file controlled by a regular user. The drawback is that it would give the regular user access to all the NGINX configuration options, while I only need to set up redirects.

To work around that I could maybe write a piece of automation which reads a file uploaded by the user, containing only information required for redirects. Something like the _redirects file in Netlify. But this feels… wrong. Too custom.

To recap:

• Static site
• HTTP 302 or HTTP 307 needed
• Frequent updates (manual editing of /etc/nginx not feasible)
• Files uploaded as an unprivileged system user
• Not giving that user too much power over the web server

What would be a good way of doing this on NGINX?

I'm setting up an Ubuntu NFS server with a Solaris 10 client. The basic configuration looks okay to me, and it was also working for some time. I'm getting an "RPC: Authentication error" message on the client.

server

/etc/exports:

/export/opencsw-future 192.168.3.0/24(rw,nohide,insecure,no_subtree_check,async)
/export/opencsw-current 192.168.3.0/24(rw,nohide,insecure,no_subtree_check,async)

$ ls -ld /export/opencsw-current
drwxr-xr-x 7 maciej maciej 4096 2012-02-05 14:55 /export/opencsw-current

client

$ grep opencsw /etc/vfstab
foosrv:/opencsw-current - /export/opencsw-current nfs - yes -
$ sudo mount /export/opencsw-current
NFS compound failed for server foosrv: error 7 (RPC: Authentication error)
(...repeated...)
nfs mount: mount: /export/opencsw-current: Permission denied

My server host name resolves to both IPv4 and IPv6 addresses.

One of the recent questions asked about potential reasons to use Solaris on a server. I would like to ask about the opposite:

What would be reasons to stay away from Solaris on a server?

Package and patch management comes to mind, but I'm sure there's more to it. You can also point out advantages of other server operating systems over Solaris.