Fetching from urls: Error Message (#16434)

* Fetching from urls: Error Message

Fix the error message when fetching from consecutive `urls` of a
package version. Each fail should show the currently failing URL,
not the first url.

Example multi-problem run that occured in real life:
```
==> 5821: Installing util-macros
curl: (28) Connection timed out after 10000 milliseconds
curl: (16) Error in the HTTP2 framing layer
curl: (22) The requested URL returned error: 403 Forbidden
==> Fetching https://www.x.org/archive/individual/util/util-macros-1.19.1.tar.bz2
==> Failed to fetch file from URL: https://www.x.org/archive/individual/util/util-macros-1.19.1.tar.bz2
    Curl failed with error 28
==> Fetching https://mirrors.ircam.fr/pub/x.org/individual/util/util-macros-1.19.1.tar.bz2
==> Failed to fetch file from URL: https://www.x.org/archive/individual/util/util-macros-1.19.1.tar.bz2
    Curl failed with error 16
==> Fetching http://xorg.mirrors.pair.com/individual/util/util-macros-1.19.1.tar.bz2
==> Failed to fetch file from URL: https://www.x.org/archive/individual/util/util-macros-1.19.1.tar.bz2
    URL https://www.x.org/archive/individual/util/util-macros-1.19.1.tar.bz2 was not found!
==> Fetching from https://www.x.org/archive/individual/util/util-macros-1.19.1.tar.bz2 failed.
==> Error: FetchError: All fetchers failed for spack-stage-util-macros-1.19.1-se2a2e74oyusj2r4esgcb7pr3qhh45ef
```

- `urls[0]`: HTTP2 layer error
- `urls[1]`: timeout
- `urls[2]`: missing file on mirror

* x.org: two more mirrors

x.org mirrors are a bit tricky, since many are out-of-sync or off.
A good package to test with is `util-macros`, which had a "recent"
release.

lib/spack/spack/build_systems/xorg.py

@@ -14,9 +14,14 @@ class XorgPackage(spack.package.PackageBase):
     xorg_mirror_path = None
 
     #: List of x.org mirrors used by Spack
+    #  Note: x.org mirrors are a bit tricky, since many are out-of-sync or off.
+    #        A good package to test with is `util-macros`, which had a "recent"
+    #        release.
     base_mirrors = [
         'https://www.x.org/archive/individual/',
         'https://mirrors.ircam.fr/pub/x.org/individual/',
+        'https://mirror.transip.net/xorg/individual/',
+        'ftp://ftp.freedesktop.org/pub/xorg/individual/',
         'http://xorg.mirrors.pair.com/individual/'
     ]
 

lib/spack/spack/fetch_strategy.py

@@ -292,6 +292,7 @@ def fetch(self):
             tty.msg("Already downloaded %s" % self.archive_file)
             return
 
+        url = None
         for url in self.candidate_urls:
             try:
                 partial_file, save_file = self._fetch_from_url(url)
@@ -303,7 +304,7 @@ def fetch(self):
                 pass
 
         if not self.archive_file:
-            raise FailedDownloadError(self.url)
+            raise FailedDownloadError(url)
 
     def _fetch_from_url(self, url):
         save_file = None
@@ -369,12 +370,12 @@ def _fetch_from_url(self, url):
             if curl.returncode == 22:
                 # This is a 404.  Curl will print the error.
                 raise FailedDownloadError(
-                    self.url, "URL %s was not found!" % self.url)
+                    url, "URL %s was not found!" % url)
 
             elif curl.returncode == 60:
                 # This is a certificate error.  Suggest spack -k
                 raise FailedDownloadError(
-                    self.url,
+                    url,
                     "Curl was unable to fetch due to invalid certificate. "
                     "This is either an attack, or your cluster's SSL "
                     "configuration is bad.  If you believe your SSL "
@@ -386,7 +387,7 @@ def _fetch_from_url(self, url):
                 # This is some other curl error.  Curl will print the
                 # error, but print a spack message too
                 raise FailedDownloadError(
-                    self.url,
+                    url,
                     "Curl failed with error %d" % curl.returncode)
 
         # Check if we somehow got an HTML file rather than the archive we