Just to confirm it also happens on Mac OS X (10.6.8) with fossil 1.24 the whole
time. This time I was able to capture the text from the initial pull since the
latest checkout.
Bytes Cards Artifacts Deltas
Sent: 146 2 0 0
fossil: *** time skew *** server is slow by 42.8 seconds
Received: 78 2 0 0
Sent: 177 2 0 0
fossil: *** time skew *** server is slow by 43.3 seconds
Received:4464 3 1 0
Sent: 647 12 0 0
Received: 3458142 13 4 7
Total network traffic: 1401 bytes sent, 1117874 bytes received
Clocking in at 1117874 bytes it looks like it is in fact grabbing the new
artifacts, but the resulting repository is 8819712 bytes.
$ ls -al
-rw-r--r-- 1 edwardblake staff 8819712 23 Jan 06:47 TortoiseFossil.fossil
I have completely forgotten to backup the fossil repo before doing this
operation however.
Opening the repository brings up the same issue.
project-code: ae039b7f3dccab4daa13783935c22c1cec6fbedc
checkout: bb5431489fc1f5cd5eca5e8a39b855d4c549d06f 2013-01-14 21:28:08 UTC
parent: b2e5f4eca7f3bbe0d7cb52932e942e9c6bed6435 2013-01-14 03:07:18 UTC
tags: trunk
comment: Added manifest for modern Windows appearance (user: edwardlblake)
--- edward.bl...@rovoscape.com wrote:
From: "Edward Blake"
To:
Subject: [fossil-users] Pull isn't pulling, or something
Date: Wed, 23 Jan 2013 02:12:13 -0800
Hello all,
I've stumbled on an issue where a fossil repository apparently is in a state
that it can't pull new checkouts from the remote url. It seemed to be working
during a "fossil all pull", at least I think it was downloading new bytes at
that time, but then the last checkout locally isn't the last checkout on the
remote when inspected with fossil ui and fossil status. I've performed a pull
with version 1.23 windows build and then when I saw the issue, I tried
upgrading to 1.24 + rebuild it and then pull again to see if that helps
(hasn't).
Performing a fossil rebuild and then fossil update shows a brief autosync
session:
Bytes Cards Artifacts Deltas
Sent: 130 1 0 0
fossil: *** time skew *** server is slow by 34.7 seconds
Received: 78 2 0 0
Total network traffic: 367 bytes sent, 402 bytes received
---
checkout: bb5431489fc1f5cd5eca5e8a39b855d4c549d06f 2013-01-14 21:28:08 UTC
tags: trunk
comment: Added manifest for modern Windows appearance (user: edwardlblake)
changes: None. Already up-to-date
In between the new checkout (which isn't being updated to) and the last
checkout I've added a bunch of binaries in the megabytes range. Cloning to a
new repository on the same machine from the remote url was no problem by the
way, it gets the latest checkout, so it's not so bad. The newly cloned repo is
12,593KB and the affected repository is 7,963KB so it seems the artifacts
aren't just stored yet inaccessible. Both have 1 leaf, 1 reportfmt, 12 tag in
fossil sqlite. From fossil sqlite here are the numbers that differ:
Fresh Affected
-
blob793788
delta 340331
event 39 38
filename456449
mlink 774760
plink 30 29
unclustered 332327
tagxref 74 72
0 rows in both: attachment, backlink, concealed, orphan, phantom, private,
shun, unsent
Since its an open source project I can email the affected local repository
file, remote repository and the originating local repository file containing
the latest commit if that helps.
Thanks,
Edward Blake
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
___
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users