Codesign not working "Timestamp service not available"

Joined
Mar 6, 2013
Messages
5
Reaction score
0
Points
1
I am new to the forums, so please be patient with me. I am having an issue with an app that we are working on and appreciate any suggestions or assistance.

We have multiple build machines that we are running, all but 1 are running 10.8. The other one is running 10.7.4.

When we codesign the app on the 10.7.4 system, we have no issues. When we attempt to codesign on the 10.8 systems, we get the error "Timestamp service is not available". The odd thing is that this only started happening about a couple of weeks ago. I do not believe this is a firewall issue, as I have two machines sitting beside me and are running the codesign from my location (10.7.4 machine and a 10.8 machine).

I am running the command: codesign -s "Developer ID Applciation: <company>" /path/to/app.

Now I know there is a workaround: --timestamp=none, but that is not what we are looking to use. I would rather fix the issue than just use a workaround, that most people admit is not a good fix. Also xcode is up to date to the latest version on both machines.

Again, I appreciate any assistance!
D
 

Raz0rEdge

Well-known member
Staff member
Moderator
Joined
Jul 17, 2009
Messages
15,745
Reaction score
2,071
Points
113
Location
MA
Your Mac's Specs
2022 Mac Studio M1 Max, 2023 M2 MBA
Welcome to Mac-Forums

Can you check to see that both Macs have a consistent connection to Apple's time server..so try something like "ping time.apple.com" and see if you notice any packet loss on either machine.

Is the issue with the Mac running 10.8 persistent, i.e., does it ever succeed or does it always fail?
 
OP
D
Joined
Mar 6, 2013
Messages
5
Reaction score
0
Points
1
If I attempt to ping time.apple.com on both machines I get:
PING time.apple.com (17.151.16.23): 56 data bytes
Request timeout for icmp_seq 0
Request timeout for icmp_seq 1
Request timeout for icmp_seq 2
Request timeout for icmp_seq 3
Request timeout for icmp_seq 4
^C
--- time.apple.com ping statistics ---
6 packets transmitted, 0 packets received, 100.0% packet loss

So I am getting packet loss. However, it does not explain why on the 10.7.4 machine we are getting the app signed.

As for your other question, it worked on the 10.8 machine up until recently. I would say in the last 2 to 3 weeks is when it stopped signing on the 10.8 machine. So it did work at one point, it just now is a consistent error.
 

Raz0rEdge

Well-known member
Staff member
Moderator
Joined
Jul 17, 2009
Messages
15,745
Reaction score
2,071
Points
113
Location
MA
Your Mac's Specs
2022 Mac Studio M1 Max, 2023 M2 MBA
OK..looks like ICMP support is off on that server, which is to be expected on most sites that accessed often..

Can you try "traceroute time.apple.com" instead on both the machines to see if the path is similar and actually reaches the destination..based on the ping output, I'd expect that you're getting through..

But at least this proves that is indeed not a firewall/network issue..

A quick search of this error indicated that it's happened before and for extended periods of time with the no timestamp workaround used to keep going for a while. However, since your 10.7.x machine is fine, that doesn't seem to be the case right now.

Are both the 10.7.x and 10.8 machine running the same version of XCode and tools?

Can you re-run your 'codesign' command prefix it with the dtruss command like so: "dtruss codesign ...."

Compare the outputs on the two computers to see if they are doing essentially the same thing..
 
OP
D
Joined
Mar 6, 2013
Messages
5
Reaction score
0
Points
1
So I used the command dtruss before I run the codesign command. And I found some interesting results.

On the 10.7.x machine, it ran through no issues. On the 10.8 machine, it started fine, however this is the output I get in some spots...which I am assuming is the issue or an issue:

getpid(0x7FFF558EB8A8, 0x10A31A004, 0xEA60) = 1007 0
__semwait_signal(0x0, 0x0, 0x1) = -1 Err#22
__semwait_signal(0x0, 0x0, 0x1) = -1 Err#22
__semwait_signal(0x0, 0x0, 0x1) = -1 Err#22
__semwait_signal(0x0, 0x0, 0x1) = -1 Err#22
__semwait_signal(0x0, 0x0, 0x1) = -1 Err#22
__semwait_signal(0x0, 0x0, 0x1) = -1 Err#22
__semwait_signal(0x0, 0x0, 0x1) = -1 Err#22
__semwait_signal(0x0, 0x0, 0x1) = -1 Err#22
__semwait_signal(0x0, 0x0, 0x1) = -1 Err#22
__semwait_signal(0x0, 0x0, 0x1) = -1 Err#22
kill (0x3E8, 0x0, 0x1) = 0 0

And this will go on for a while, run some items, then try to getpid and do the semwait and kill again for a bit. Quite odd.

Any thoughts?
 

Raz0rEdge

Well-known member
Staff member
Moderator
Joined
Jul 17, 2009
Messages
15,745
Reaction score
2,071
Points
113
Location
MA
Your Mac's Specs
2022 Mac Studio M1 Max, 2023 M2 MBA
A few guesses on my part..

Looks like the program is trying to find the PID of another process and has a semaphore associated with that action that is most likely timing out. So whatever it is looking for doesn't exist..

Do you see similar getpid() commands on the working 10.7.x machine?
 
OP
D
Joined
Mar 6, 2013
Messages
5
Reaction score
0
Points
1
I have tested it on multiple machines and that getpid() occurs on all of them, the 10.8 and the 10.7.x machine.

I appreciate the assistance. We are trying to get a ticket open with Apple to take a look.
 

Raz0rEdge

Well-known member
Staff member
Moderator
Joined
Jul 17, 2009
Messages
15,745
Reaction score
2,071
Points
113
Location
MA
Your Mac's Specs
2022 Mac Studio M1 Max, 2023 M2 MBA
OK, it sounds like Apple might be able to give you a better answer then my half guesses..:) Also since I haven't used this application in particular, so I can't think of any other guesses..
 

Shop Amazon


Shop for your Apple, Mac, iPhone and other computer products on Amazon.
We are a participant in the Amazon Services LLC Associates Program, an affiliate program designed to provide a means for us to earn fees by linking to Amazon and affiliated sites.
Top