mgr/cephadm/inventory: do not try to resolve current mgr host

The CNI configuration may set up a private network for the container, which
is mapped to the hostname in /etc/hosts.  For example, my test box sets
up 10.88.0.0/24 because I was using crio + kubeadm on this host earlier
(at least I think that's why):

$ sudo podman run --rm --name test123 --entrypoint /bin/bash -it quay.ceph.io/ceph-ci/ceph:master -c "cat /etc/hosts"
127.0.0.1   localhost localhost.localdomain localhost4 localhost4.localdomain4
::1         localhost localhost.localdomain localhost6 localhost6.localdomain6
10.88.0.8	f9e91bf2478f test123

In any case, we should never trust a lookup of our own hostname from inside
a container!

This isn't quite sufficient, though: if this is a single-host cluster, then
we fall back to using get_mgr_ip(). That value may be distorted by the
public_network option on the mgr, but we don't have any other good
options here, and single-node clusters are unlikely to have complex
network configs.

Refactor a bit to avoid the try/except nesting.

Signed-off-by: Sage Weil <sage@newdream.net>

diff --git a/src/pybind/mgr/cephadm/inventory.py b/src/pybind/mgr/cephadm/inventory.py
index c2ce50db8c86b929179b0a199e87677ac825e1ab..d0509b66124eea5118c764feb609f52fc1c4f271 100644 (file)
--- a/src/pybind/mgr/cephadm/inventory.py
+++ b/src/pybind/mgr/cephadm/inventory.py
@@ -3,6 +3,7 @@ from copy import copy
 import ipaddress
 import json
 import logging
+import socket
 from typing import TYPE_CHECKING, Dict, List, Iterator, Optional, Any, Tuple, Set, Mapping, cast, \
     NamedTuple, Type
 
@@ -32,6 +33,14 @@ class Inventory:
     def __init__(self, mgr: 'CephadmOrchestrator'):
         self.mgr = mgr
         adjusted_addrs = False
+
+        def is_valid_ip(ip: str) -> bool:
+            try:
+                ipaddress.ip_address(ip)
+                return True
+            except ValueError:
+                return False
+
         # load inventory
         i = self.mgr.get_store('inventory')
         if i:
@@ -42,18 +51,30 @@ class Inventory:
                     v['hostname'] = k
 
                 # convert legacy non-IP addr?
-                try:
-                    ipaddress.ip_address(v.get('addr'))
-                except ValueError:
+                if is_valid_ip(str(v.get('addr'))):
+                    continue
+                if len(self._inventory) > 1:
+                    if k == socket.gethostname():
+                        # Never try to resolve our own host!  This is
+                        # fraught and can lead to either a loopback
+                        # address (due to podman's futzing with
+                        # /etc/hosts) or a private IP based on the CNI
+                        # configuration.  Instead, wait until the mgr
+                        # fails over to another host and let them resolve
+                        # this host.
+                        continue
                     ip = resolve_ip(cast(str, v.get('addr')))
-                    try:
-                        ipaddress.ip_address(ip)
-                        if not ip.startswith('127.0.'):
-                            self.mgr.log.info(f"inventory: adjusted host {v['hostname']} addr '{v['addr']}' -> '{ip}'")
-                            v['addr'] = ip
-                            adjusted_addrs = True
-                    except ValueError:
-                        pass
+                else:
+                    # we only have 1 node in the cluster, so we can't
+                    # rely on another host doing the lookup.  use the
+                    # IP the mgr binds to.
+                    ip = self.mgr.get_mgr_ip()
+                if is_valid_ip(ip) and not ip.startswith('127.0.'):
+                    self.mgr.log.info(
+                        f"inventory: adjusted host {v['hostname']} addr '{v['addr']}' -> '{ip}'"
+                    )
+                    v['addr'] = ip
+                    adjusted_addrs = True
             if adjusted_addrs:
                 self.save()
         else: