You are viewing a plain text version of this content. The canonical link for it is here.
Posted to repo-maintainers@maven.apache.org by Carlos Sanchez <ca...@apache.org> on 2010/01/21 01:50:24 UTC
Re: Attn Brian Fox (Re: [Maven Central Repository Synchronization]
FAILURE)
yeah, you need to subscribe
the traceroute is stopping at
8 2120.tengigabitethernet-0-0-0.lns01.tor.packetflow.ca
(69.196.136.79) 36.212 ms 37.496 ms 35.958 ms
9 * * *
On Wed, Jan 20, 2010 at 4:43 PM, Owen Jacobson
<ow...@grimoire.ca> wrote:
> On Jan 20, 2010, at 7:01 PM, maven@repo1.maven.org wrote:
>
>> --- Some repositories were not synchronized ---
>> groupId: ca.grimoire
>> Error:
>> rsync: failed to connect to alchemy.grimoire.ca: No route to host (113)
>> rsync error: error in socket IO (code 10) at clientserver.c(122) [receiver=3.0.3]
>> Error synchronizing metadata. Exit code: 10
>> Command line executed: /bin/sh -c "rsync --include=*/ --include=**/maven-metadata.xml* --exclude=* --exclude-from=/home/maven/bin/synchronize/syncopate/exclusions.txt -Lrtivz rsync://alchemy.grimoire.ca/m2/releases/ca/grimoire/ /home/maven/repository-staging/to-ibiblio/maven2/ca/grimoire/"
>
> On Tue Jan 19, 2010, at 2:11 AM, Brian Fox wrote:
>
>> Ping is working from Central, so lets see if the next sync is ok.
>
> Nope, looks like it isn't.
>
> Let me know if you need anything from me on this -- please CC me on replies, too, since I don't subscribe to repo-maintainers (but I probably should).
>
> -o
>
>
Re: Grimoire (Re: [Maven Central Repository Synchronization] FAILURE)
Posted by Owen Jacobson <ow...@grimoire.ca>.
On Jan 20, 2010, at 8:14 PM, Carlos Sanchez wrote:
> $ dig NS grimoire.ca
...
> ;; QUESTION SECTION:
> ;grimoire.ca. IN NS
>
> ;; ANSWER SECTION:
> grimoire.ca. 84866 IN NS ns1.geekfire.com.
> grimoire.ca. 84866 IN NS ns2.geekfire.com.
So not only is it getting the old name servers, but it got them *again* very recently. Surreal. I'll get in touch with the guy who runs those servers and tell him to refresh the zone, just in case that's where it's coming from.
> ;; SERVER: 63.246.7.201#53(63.246.7.201)
Any idea why this server is still remembering stale NS records?
-o
Re: Attn Brian Fox (Re: [Maven Central Repository Synchronization]
FAILURE)
Posted by Carlos Sanchez <ca...@apache.org>.
$ dig NS grimoire.ca
; <<>> DiG 9.2.4 <<>> NS grimoire.ca
;; global options: printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 56286
;; flags: qr rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 0
;; QUESTION SECTION:
;grimoire.ca. IN NS
;; ANSWER SECTION:
grimoire.ca. 84866 IN NS ns1.geekfire.com.
grimoire.ca. 84866 IN NS ns2.geekfire.com.
;; Query time: 25 msec
;; SERVER: 63.246.7.201#53(63.246.7.201)
;; WHEN: Wed Jan 20 19:13:58 2010
;; MSG SIZE rcvd: 77
$ dig +norecurse @a.ca-servers.ca NS grimoire.ca
; <<>> DiG 9.2.4 <<>> +norecurse @a.ca-servers.ca NS grimoire.ca
; (1 server found)
;; global options: printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 18476
;; flags: qr; QUERY: 1, ANSWER: 0, AUTHORITY: 5, ADDITIONAL: 0
;; QUESTION SECTION:
;grimoire.ca. IN NS
;; AUTHORITY SECTION:
grimoire.ca. 86400 IN NS ns3.linode.com.
grimoire.ca. 86400 IN NS ns2.linode.com.
grimoire.ca. 86400 IN NS ns5.linode.com.
grimoire.ca. 86400 IN NS ns1.linode.com.
grimoire.ca. 86400 IN NS ns4.linode.com.
;; Query time: 29 msec
;; SERVER: 192.228.27.11#53(192.228.27.11)
;; WHEN: Wed Jan 20 19:13:58 2010
;; MSG SIZE rcvd: 129
$ dig +norecurse @ns1.linode.com NS grimoire.ca
; <<>> DiG 9.2.4 <<>> +norecurse @ns1.linode.com NS grimoire.ca
; (1 server found)
;; global options: printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 65500
;; flags: qr aa; QUERY: 1, ANSWER: 5, AUTHORITY: 0, ADDITIONAL: 5
;; QUESTION SECTION:
;grimoire.ca. IN NS
;; ANSWER SECTION:
grimoire.ca. 3600 IN NS ns5.linode.com.
grimoire.ca. 3600 IN NS ns2.linode.com.
grimoire.ca. 3600 IN NS ns3.linode.com.
grimoire.ca. 3600 IN NS ns1.linode.com.
grimoire.ca. 3600 IN NS ns4.linode.com.
;; ADDITIONAL SECTION:
ns1.linode.com. 86400 IN A 69.93.127.10
ns2.linode.com. 86400 IN A 65.19.178.10
ns3.linode.com. 86400 IN A 75.127.96.10
ns4.linode.com. 86400 IN A 207.192.70.10
ns5.linode.com. 86400 IN A 109.74.194.10
;; Query time: 34 msec
;; SERVER: 69.93.127.10#53(69.93.127.10)
;; WHEN: Wed Jan 20 19:14:01 2010
;; MSG SIZE rcvd: 209
On Wed, Jan 20, 2010 at 5:06 PM, Owen Jacobson
<ow...@grimoire.ca> wrote:
> Ugh, thanks. Can you fire me the output of each of
>
> dig NS grimoire.ca
> dig +norecurse @a.ca-servers.ca NS grimoire.ca
> dig +norecurse @ns1.linode.com NS grimoire.ca
>
> from that machine so that I can identify which part of the delegation or expiry is hosed?
>
> -o
>
> On Jan 20, 2010, at 8:03 PM, Carlos Sanchez wrote:
>
>> FYI is not propagated in the repo machine, although it is in my laptop
>>
>> $ dig alchemy.grimoire.ca
>>
>> ; <<>> DiG 9.2.4 <<>> alchemy.grimoire.ca
>> ;; global options: printcmd
>> ;; Got answer:
>> ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 53285
>> ;; flags: qr rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 0
>>
>> ;; QUESTION SECTION:
>> ;alchemy.grimoire.ca. IN A
>>
>> ;; ANSWER SECTION:
>> alchemy.grimoire.ca. 45446 IN CNAME public.grimoire.ca.
>> public.grimoire.ca. 2874 IN A 76.10.176.194
>>
>> ;; Query time: 29 msec
>> ;; SERVER: 63.246.7.201#53(63.246.7.201)
>> ;; WHEN: Wed Jan 20 19:00:30 2010
>> ;; MSG SIZE rcvd: 74
>>
>>
>>
>> On Wed, Jan 20, 2010 at 4:58 PM, Owen Jacobson
>> <ow...@grimoire.ca> wrote:
>>> It shouldn't be going anywhere near tor.packetflow.ca any more.
>>>
>>> On Sunday I updated the DNS entry for alchemy.grimoire.ca from 76.10.176.194 (its old home, connected to the internet via Teksavvy and packetflow.ca) to 69.164.211.26 (its new home, connected via nac.net). I can't find any DNS servers that are still caching the old IP address, but the expiry time for that entry was set to 86400 seconds, which has now long since elapsed. The only thing that I can think of is that something is hosed in my (well, mine or Linode's) DNS setup, preventing the synch script or the resolver on its host from noticing the changeover.
>>>
>>> Thoughts? Observations?
>>>
>>> (I am now subscribed to repo-maintainers. :)
>>>
>>> -o
>>>
>>> On Jan 20, 2010, at 7:50 PM, Carlos Sanchez wrote:
>>>
>>>> yeah, you need to subscribe
>>>>
>>>> the traceroute is stopping at
>>>> 8 2120.tengigabitethernet-0-0-0.lns01.tor.packetflow.ca
>>>> (69.196.136.79) 36.212 ms 37.496 ms 35.958 ms
>>>> 9 * * *
>>>>
>>>> On Wed, Jan 20, 2010 at 4:43 PM, Owen Jacobson
>>>> <ow...@grimoire.ca> wrote:
>>>>> On Jan 20, 2010, at 7:01 PM, maven@repo1.maven.org wrote:
>>>>>
>>>>>> --- Some repositories were not synchronized ---
>>>>>> groupId: ca.grimoire
>>>>>> Error:
>>>>>> rsync: failed to connect to alchemy.grimoire.ca: No route to host (113)
>>>>>> rsync error: error in socket IO (code 10) at clientserver.c(122) [receiver=3.0.3]
>>>>>> Error synchronizing metadata. Exit code: 10
>>>>>> Command line executed: /bin/sh -c "rsync --include=*/ --include=**/maven-metadata.xml* --exclude=* --exclude-from=/home/maven/bin/synchronize/syncopate/exclusions.txt -Lrtivz rsync://alchemy.grimoire.ca/m2/releases/ca/grimoire/ /home/maven/repository-staging/to-ibiblio/maven2/ca/grimoire/"
>>>>>
>>>>> On Tue Jan 19, 2010, at 2:11 AM, Brian Fox wrote:
>>>>>
>>>>>> Ping is working from Central, so lets see if the next sync is ok.
>>>>>
>>>>> Nope, looks like it isn't.
>>>>>
>>>>> Let me know if you need anything from me on this -- please CC me on replies, too, since I don't subscribe to repo-maintainers (but I probably should).
>>>
>>>
>>>
>
>
Re: Attn Brian Fox (Re: [Maven Central Repository Synchronization] FAILURE)
Posted by Owen Jacobson <ow...@grimoire.ca>.
Ugh, thanks. Can you fire me the output of each of
dig NS grimoire.ca
dig +norecurse @a.ca-servers.ca NS grimoire.ca
dig +norecurse @ns1.linode.com NS grimoire.ca
from that machine so that I can identify which part of the delegation or expiry is hosed?
-o
On Jan 20, 2010, at 8:03 PM, Carlos Sanchez wrote:
> FYI is not propagated in the repo machine, although it is in my laptop
>
> $ dig alchemy.grimoire.ca
>
> ; <<>> DiG 9.2.4 <<>> alchemy.grimoire.ca
> ;; global options: printcmd
> ;; Got answer:
> ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 53285
> ;; flags: qr rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 0
>
> ;; QUESTION SECTION:
> ;alchemy.grimoire.ca. IN A
>
> ;; ANSWER SECTION:
> alchemy.grimoire.ca. 45446 IN CNAME public.grimoire.ca.
> public.grimoire.ca. 2874 IN A 76.10.176.194
>
> ;; Query time: 29 msec
> ;; SERVER: 63.246.7.201#53(63.246.7.201)
> ;; WHEN: Wed Jan 20 19:00:30 2010
> ;; MSG SIZE rcvd: 74
>
>
>
> On Wed, Jan 20, 2010 at 4:58 PM, Owen Jacobson
> <ow...@grimoire.ca> wrote:
>> It shouldn't be going anywhere near tor.packetflow.ca any more.
>>
>> On Sunday I updated the DNS entry for alchemy.grimoire.ca from 76.10.176.194 (its old home, connected to the internet via Teksavvy and packetflow.ca) to 69.164.211.26 (its new home, connected via nac.net). I can't find any DNS servers that are still caching the old IP address, but the expiry time for that entry was set to 86400 seconds, which has now long since elapsed. The only thing that I can think of is that something is hosed in my (well, mine or Linode's) DNS setup, preventing the synch script or the resolver on its host from noticing the changeover.
>>
>> Thoughts? Observations?
>>
>> (I am now subscribed to repo-maintainers. :)
>>
>> -o
>>
>> On Jan 20, 2010, at 7:50 PM, Carlos Sanchez wrote:
>>
>>> yeah, you need to subscribe
>>>
>>> the traceroute is stopping at
>>> 8 2120.tengigabitethernet-0-0-0.lns01.tor.packetflow.ca
>>> (69.196.136.79) 36.212 ms 37.496 ms 35.958 ms
>>> 9 * * *
>>>
>>> On Wed, Jan 20, 2010 at 4:43 PM, Owen Jacobson
>>> <ow...@grimoire.ca> wrote:
>>>> On Jan 20, 2010, at 7:01 PM, maven@repo1.maven.org wrote:
>>>>
>>>>> --- Some repositories were not synchronized ---
>>>>> groupId: ca.grimoire
>>>>> Error:
>>>>> rsync: failed to connect to alchemy.grimoire.ca: No route to host (113)
>>>>> rsync error: error in socket IO (code 10) at clientserver.c(122) [receiver=3.0.3]
>>>>> Error synchronizing metadata. Exit code: 10
>>>>> Command line executed: /bin/sh -c "rsync --include=*/ --include=**/maven-metadata.xml* --exclude=* --exclude-from=/home/maven/bin/synchronize/syncopate/exclusions.txt -Lrtivz rsync://alchemy.grimoire.ca/m2/releases/ca/grimoire/ /home/maven/repository-staging/to-ibiblio/maven2/ca/grimoire/"
>>>>
>>>> On Tue Jan 19, 2010, at 2:11 AM, Brian Fox wrote:
>>>>
>>>>> Ping is working from Central, so lets see if the next sync is ok.
>>>>
>>>> Nope, looks like it isn't.
>>>>
>>>> Let me know if you need anything from me on this -- please CC me on replies, too, since I don't subscribe to repo-maintainers (but I probably should).
>>
>>
>>
Re: Attn Brian Fox (Re: [Maven Central Repository Synchronization]
FAILURE)
Posted by Carlos Sanchez <ca...@apache.org>.
FYI is not propagated in the repo machine, although it is in my laptop
$ dig alchemy.grimoire.ca
; <<>> DiG 9.2.4 <<>> alchemy.grimoire.ca
;; global options: printcmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 53285
;; flags: qr rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 0
;; QUESTION SECTION:
;alchemy.grimoire.ca. IN A
;; ANSWER SECTION:
alchemy.grimoire.ca. 45446 IN CNAME public.grimoire.ca.
public.grimoire.ca. 2874 IN A 76.10.176.194
;; Query time: 29 msec
;; SERVER: 63.246.7.201#53(63.246.7.201)
;; WHEN: Wed Jan 20 19:00:30 2010
;; MSG SIZE rcvd: 74
On Wed, Jan 20, 2010 at 4:58 PM, Owen Jacobson
<ow...@grimoire.ca> wrote:
> It shouldn't be going anywhere near tor.packetflow.ca any more.
>
> On Sunday I updated the DNS entry for alchemy.grimoire.ca from 76.10.176.194 (its old home, connected to the internet via Teksavvy and packetflow.ca) to 69.164.211.26 (its new home, connected via nac.net). I can't find any DNS servers that are still caching the old IP address, but the expiry time for that entry was set to 86400 seconds, which has now long since elapsed. The only thing that I can think of is that something is hosed in my (well, mine or Linode's) DNS setup, preventing the synch script or the resolver on its host from noticing the changeover.
>
> Thoughts? Observations?
>
> (I am now subscribed to repo-maintainers. :)
>
> -o
>
> On Jan 20, 2010, at 7:50 PM, Carlos Sanchez wrote:
>
>> yeah, you need to subscribe
>>
>> the traceroute is stopping at
>> 8 2120.tengigabitethernet-0-0-0.lns01.tor.packetflow.ca
>> (69.196.136.79) 36.212 ms 37.496 ms 35.958 ms
>> 9 * * *
>>
>> On Wed, Jan 20, 2010 at 4:43 PM, Owen Jacobson
>> <ow...@grimoire.ca> wrote:
>>> On Jan 20, 2010, at 7:01 PM, maven@repo1.maven.org wrote:
>>>
>>>> --- Some repositories were not synchronized ---
>>>> groupId: ca.grimoire
>>>> Error:
>>>> rsync: failed to connect to alchemy.grimoire.ca: No route to host (113)
>>>> rsync error: error in socket IO (code 10) at clientserver.c(122) [receiver=3.0.3]
>>>> Error synchronizing metadata. Exit code: 10
>>>> Command line executed: /bin/sh -c "rsync --include=*/ --include=**/maven-metadata.xml* --exclude=* --exclude-from=/home/maven/bin/synchronize/syncopate/exclusions.txt -Lrtivz rsync://alchemy.grimoire.ca/m2/releases/ca/grimoire/ /home/maven/repository-staging/to-ibiblio/maven2/ca/grimoire/"
>>>
>>> On Tue Jan 19, 2010, at 2:11 AM, Brian Fox wrote:
>>>
>>>> Ping is working from Central, so lets see if the next sync is ok.
>>>
>>> Nope, looks like it isn't.
>>>
>>> Let me know if you need anything from me on this -- please CC me on replies, too, since I don't subscribe to repo-maintainers (but I probably should).
>
>
>
Re: Attn Brian Fox (Re: [Maven Central Repository Synchronization] FAILURE)
Posted by Owen Jacobson <ow...@grimoire.ca>.
It shouldn't be going anywhere near tor.packetflow.ca any more.
On Sunday I updated the DNS entry for alchemy.grimoire.ca from 76.10.176.194 (its old home, connected to the internet via Teksavvy and packetflow.ca) to 69.164.211.26 (its new home, connected via nac.net). I can't find any DNS servers that are still caching the old IP address, but the expiry time for that entry was set to 86400 seconds, which has now long since elapsed. The only thing that I can think of is that something is hosed in my (well, mine or Linode's) DNS setup, preventing the synch script or the resolver on its host from noticing the changeover.
Thoughts? Observations?
(I am now subscribed to repo-maintainers. :)
-o
On Jan 20, 2010, at 7:50 PM, Carlos Sanchez wrote:
> yeah, you need to subscribe
>
> the traceroute is stopping at
> 8 2120.tengigabitethernet-0-0-0.lns01.tor.packetflow.ca
> (69.196.136.79) 36.212 ms 37.496 ms 35.958 ms
> 9 * * *
>
> On Wed, Jan 20, 2010 at 4:43 PM, Owen Jacobson
> <ow...@grimoire.ca> wrote:
>> On Jan 20, 2010, at 7:01 PM, maven@repo1.maven.org wrote:
>>
>>> --- Some repositories were not synchronized ---
>>> groupId: ca.grimoire
>>> Error:
>>> rsync: failed to connect to alchemy.grimoire.ca: No route to host (113)
>>> rsync error: error in socket IO (code 10) at clientserver.c(122) [receiver=3.0.3]
>>> Error synchronizing metadata. Exit code: 10
>>> Command line executed: /bin/sh -c "rsync --include=*/ --include=**/maven-metadata.xml* --exclude=* --exclude-from=/home/maven/bin/synchronize/syncopate/exclusions.txt -Lrtivz rsync://alchemy.grimoire.ca/m2/releases/ca/grimoire/ /home/maven/repository-staging/to-ibiblio/maven2/ca/grimoire/"
>>
>> On Tue Jan 19, 2010, at 2:11 AM, Brian Fox wrote:
>>
>>> Ping is working from Central, so lets see if the next sync is ok.
>>
>> Nope, looks like it isn't.
>>
>> Let me know if you need anything from me on this -- please CC me on replies, too, since I don't subscribe to repo-maintainers (but I probably should).