Rocky Linux support added (#18)

* refactoring crawler - split Ubuntu handling from updater/service.py as first to make code more readable - note: debugging code still in place Signed-off-by: Christian Otto Stelter <[email protected]> * refactoring crawler - split Debian handling from updater/service.py - split Alma Linux handling from updater/service.py Signed-off-by: Christian Otto Stelter <[email protected]> * Added support for crawling Flatcar Container Linux Signed-off-by: Christian Otto Stelter <[email protected]> * - removed no longer used release_update_check from updater/service.py - added error message for unsupported distributions - updated changelog - updated README.md with supported distributions Signed-off-by: Christian Otto Stelter <[email protected]> * - make use of loguru for simple configuration of logging - AlmaLinux crawling not yet fully functional again Signed-off-by: Christian Otto Stelter <[email protected]> * - AlmaLinux crawling improved - it does not simply fetches the first hit - and working again - extended requirements.txt for loguru Signed-off-by: Christian Otto Stelter <[email protected]> * - added Debian 12 aka bookworm to image-sources.yaml - fixed debug output for checksums Signed-off-by: Christian Otto Stelter <[email protected]> * - added first try on Fedora - template not yet finished - last checksum query must be adjusted for distributions like Fedora - exporter must be adjusted Signed-off-by: Christian Otto Stelter <[email protected]> * - adjusted checksum query for Fedora - could be more generic for Distributions with no "minor" release updates - adjusted database queries - get distribution_version needed for Fedora - updated template for export Signed-off-by: Christian Otto Stelter <[email protected]> * - added --debug as argument - added first try on Fedora Linux Support to crawler - added Debian Linux 12 aka bookworm to sample image-sources.yaml Signed-off-by: Christian Otto Stelter <[email protected]> * - removed old branch from Dockerfile Signed-off-by: Christian Otto Stelter <[email protected]> * - added Rocky Linux for crawler - added template for Rocky Linux for exporter Signed-off-by: Christian Otto Stelter <[email protected]> * - corrected template for Rocky Linux Signed-off-by: Christian Otto Stelter <[email protected]> * Updated Changelog and README Signed-off-by: Christian Otto Stelter <[email protected]> --------- Signed-off-by: Christian Otto Stelter <[email protected]> Signed-off-by: Christian Otto Stelter <[email protected]> Co-authored-by: Christian Otto Stelter <[email protected]>
pluscloudopen · Jun 12, 2023 · 2a0cc87 · 2a0cc87
1 parent dc62071
commit 2a0cc87
Show file tree

Hide file tree

Showing 7 changed files with 219 additions and 4 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -10,6 +10,7 @@
 - added --debug as argument
 - added first try on Fedora Linux Support to crawler
 - added Debian Linux 12 aka bookworm to sample image-sources.yaml
+- added Rocky Linux
 
 ## 2023-06-01
 - updated example Dockerfile to new repos

diff --git a/README.md b/README.md
@@ -9,6 +9,8 @@ Supported distributions:
 - AlmaLinux
 - Flatcar Container Linux
 - Fedora Linux
+- Rocky Linux
+
 
 Note: Flatcar Container Linux offers only zipped images, so a direct upload via OpenStack Image Manager/Glance is not supported (yet).
 

diff --git a/crawler/updater/fedora.py b/crawler/updater/fedora.py
@@ -60,7 +60,7 @@ def get_image_filename(release, images_url):
     else:
         return None
 
-def get_checksum(release, images_url, images_filename):
+def get_checksum(release, images_url, image_filename):
     request = requests.get(images_url, allow_redirects=True)
     soup = BeautifulSoup(request.text, "html.parser")
 
@@ -87,7 +87,7 @@ def get_checksum(release, images_url, images_filename):
         # skip comment starting with hash
         if re.match('^#', line):
             continue
-        if images_filename in line:
+        if image_filename in line:
             # logger.debug("matched: " + line)
             (filename, new_checksum) = line.split(" = ")
             # logger.debug("new_checksum: " + new_checksum)

diff --git a/crawler/updater/rocky.py b/crawler/updater/rocky.py
@@ -0,0 +1,152 @@
+# rocky.py
+#
+# crawl distribution Rocky Linux
+
+import requests
+import re
+
+from crawler.web.generic import url_get_last_modified, url_fetch_content
+from crawler.web.directory import web_get_current_image_metadata
+
+from bs4 import BeautifulSoup
+from loguru import logger
+
+
+def build_image_url(release, major_minor, imagefile_name):
+    if not release["baseURL"].endswith("/"):
+        base_url = release["baseURL"] + "/" + major_minor + "/"
+    else:
+        base_url = release["baseURL"] + major_minor + "/"
+
+    # if not versionpath.endswith("/"):
+    #    versionpath = versionpath + "/"
+
+    return (
+        base_url + release["imagepath"] + "/" + imagefile_name
+    )
+
+def get_metadata(release):
+    # as specified in image-sources.yaml
+    # baseURL: https://download.rockylinux.org/pub/rocky/
+    if not release["baseURL"].endswith("/"):
+        base_url = release["baseURL"] + "/" + release["name"] + "/"
+    else:
+        base_url = release["baseURL"] + release["name"] + "/"
+
+    requestURL = base_url + release["imagepath"]
+    # TODO: make this configurable in image-source.yaml
+    #
+    #       group(1) contains major version as 9
+    #       group(2) contains full version with minor version number as 9.2
+    #       group(3) contains release date as 20230513
+    #       group(4) contains release date suffix 0 (subversion?)
+    #       ex. Rocky-9-GenericCloud-Base-9.2-20230513.0.x86_64.qcow2
+
+    filename_pattern = re.compile(r"Rocky-(\d+)-GenericCloud-Base-(\d+\.\d+)-(\d+).(\d+)\.x86_64\.qcow2")
+
+    logger.debug("request_URL: " + requestURL)
+
+    request = requests.get(requestURL, allow_redirects=True)
+    soup = BeautifulSoup(request.text, "html.parser")
+
+    for link in soup.find_all("a"):
+        data = link.get("href")
+        # logger.debug("data: " + data)
+
+        if filename_pattern.search(data):
+            logger.debug("pattern matched for " + data)
+            extract = filename_pattern.search(data)
+            major_minor = extract.group(2)
+            version = extract.group(3)
+
+            release_date = (
+                version[0:4]
+                + "-"
+                + version[4:6]
+                + "-"
+                + version[6:8]
+            )
+            logger.debug("url: " + build_image_url(release, major_minor, data))
+            logger.debug("last version: " + version)
+            logger.debug("release_date: " + release_date)
+
+            return {
+                "url": build_image_url(release, major_minor, data),
+                "version": version,
+                "release_date": release_date,
+            }
+
+    return None
+
+def get_checksum(checksum_url, image_filename):
+
+    checksum_list = url_fetch_content(checksum_url)
+    if checksum_list is None:
+        return None
+
+    for line in checksum_list.splitlines():
+        # logger.debug("line: " + line)
+
+        # skip comment starting with hash
+        if re.match('^#', line):
+            continue
+        if image_filename in line:
+            # logger.debug("matched: " + line)
+            (filename, new_checksum) = line.split(" = ")
+            # logger.debug("new_checksum: " + new_checksum)
+
+            return new_checksum
+
+    return None
+
+def rocky_update_check(release, last_checksum):
+    # as specified in image-sources.yaml
+    # baseURL: https://download.rockylinux.org/pub/rocky/
+    if not release["baseURL"].endswith("/"):
+        base_url = release["baseURL"] + "/" + release["name"] + "/"
+    else:
+        base_url = release["baseURL"] + release["name"] + "/"
+
+    checksum_url = base_url + release["imagepath"] + "/" + release["checksumname"]
+
+    logger.debug("checksum_url: " + checksum_url)
+
+    # as specified in image-sources.yaml
+    # imagename: Rocky-9-GenericCloud.latest.x86_64
+    # extension: qcow2
+    imagename = release["imagename"] + "." + release["extension"]
+
+    logger.debug("imagename: " + imagename)
+
+    current_checksum = get_checksum(checksum_url, imagename)
+
+    if current_checksum is None:
+        logger.error(
+            "no matching checksum found - check image (%s) "
+            "and checksum filename (%s)" % (imagename, release["checksumname"])
+        )
+        return None
+
+    logger.debug("current_checksum: " + current_checksum)
+
+    # as specified in image-sources.yaml
+    # algorithm: sha256
+    current_checksum = release["algorithm"] + ":" + current_checksum
+
+    if current_checksum != last_checksum:
+        logger.debug("current_checksum " + current_checksum + " differs from last_checksum " + last_checksum)
+
+        image_metadata = get_metadata(release)
+        if image_metadata is not None:
+            logger.debug("got metadata")
+            update = {}
+            update["release_date"] = image_metadata["release_date"]
+            update["url"] = image_metadata["url"]
+            update["version"] = image_metadata["version"]
+            update["checksum"] = current_checksum
+            return update
+        else:
+            logger.warning("got no metadata")
+            return None
+
+    return None
diff --git a/crawler/updater/service.py b/crawler/updater/service.py
@@ -6,6 +6,8 @@
 from crawler.updater.alma import alma_update_check
 from crawler.updater.flatcar import flatcar_update_check
 from crawler.updater.fedora import fedora_update_check
+from crawler.updater.rocky import rocky_update_check
+
 
 
 def image_update_service(connection, source):
@@ -32,6 +34,9 @@ def image_update_service(connection, source):
             catalog_update = flatcar_update_check(release, last_checksum)
         elif "Fedora" in release["imagename"]:
             catalog_update = fedora_update_check(release, last_checksum)
+        elif "Rocky" in release["imagename"]:
+            catalog_update = rocky_update_check(release, last_checksum)
+
         else:
             logger.error("Unsupported distribution " + source["name"] + " - please check your images-sources.yaml")
             raise SystemExit(1)

diff --git a/etc/image-sources.yaml b/etc/image-sources.yaml
@@ -102,8 +102,7 @@ sources:
         extension: img.gz
         checksumname: flatcar_production_openstack_image.img.gz.DIGESTS
         algorithm: md5
-        immutable: true
-
+
   - name: Fedora
     vendor: "Fedora Project"
     releases:
@@ -117,3 +116,25 @@ sources:
         checksumname: CHECKSUM
         algorithm: sha256
         limit: 1
+
+  - name: RockyLinux
+    vendor: "Rocky Linux Foundation"
+    releases:
+      - name: '8'
+        codename: 'none'
+        baseURL: https://download.rockylinux.org/pub/rocky/
+        imagepath: images/x86_64
+        imagename: Rocky-8-GenericCloud.latest.x86_64
+        extension: qcow2
+        checksumname: CHECKSUM
+        algorithm: sha256
+        limit: 1
+      - name: '9'
+        codename: 'none'
+        baseURL: https://download.rockylinux.org/pub/rocky/
+        imagepath: images/x86_64
+        imagename: Rocky-9-GenericCloud.latest.x86_64
+        extension: qcow2
+        checksumname: CHECKSUM
+        algorithm: sha256
+        limit: 1
diff --git a/templates/rockylinux.yml.j2 b/templates/rockylinux.yml.j2
@@ -0,0 +1,34 @@
+
+  - name: {{ catalog['name'] }} {{ catalog['os_version'] }}
+    format: qcow2
+    login: rocky
+    min_disk: 10
+    min_ram: 512
+    status: active
+    visibility: public
+    multi: true
+    meta:
+      architecture: x86_64
+      hypervisor_type: qemu
+      hw_disk_bus: scsi
+      hw_rng_model: virtio
+      hw_scsi_model: virtio-scsi
+      hw_qemu_guest_agent: yes
+      hw_watchdog_action: reset
+      replace_frequency: quarterly
+      hotfix_hours: 0
+      uuid_validity: last-3
+      provided_until: none
+      os_distro: rocky
+      os_version: '{{ catalog['os_version'] }}'
+    tags: []
+    latest_checksum_url: {{ metadata['baseURL'] }}{{ catalog['os_version'] }}/{{ metadata['imagepath'] }}/{{ metadata['checksumname'] }}
+    latest_url: {{ metadata['baseURL'] }}{{ catalog['os_version'] }}/{{ metadata['imagepath'] }}/{{ metadata['imagename'] }}.{{ metadata['extension'] }}
+    versions:{% for release_version in catalog['versions'] %}
+      - version: '{{ release_version }}'
+        url: {{ catalog['versions'][release_version]['url'] }}
+        checksum: {{ catalog['versions'][release_version]['checksum'] }}
+        build_date: {{ catalog['versions'][release_version]['release_date'] }}
+        image_source: {{ catalog['versions'][release_version]['url'] }}
+        image_description: https://docs.rockylinux.org/release_notes/
+{%- endfor %}