Possible network issue

[Robert Engels]

Actually, given what I now know I can probably write a standalone test case that demonstrates the problem. 

I’ll see if I can do so after work tonight. 

> On Nov 8, 2023, at 6:39 AM, Robert Engels <rengels at ix.netcom.com> wrote:
> 
> 
> The thing is it never denies the memory request - just delays it. I am fairly certain if I raise the mbufs it will work - similar to lowering the transfer size - but I don’t think this is a viable solution. I think it needs to be resolved in the net layer. 
> 
> I could build the JDK “fixing” that method but that might take me a while. I have to assume you’re already set up to do so - I’m happy to test a build with the needed changes. :)
> 
>>> On Nov 8, 2023, at 2:07 AM, Alan Bateman <Alan.Bateman at oracle.com> wrote:
>>> 
>>  On 08/11/2023 00:50, robert engels wrote:
>>> iMac:helidon_test robertengels$ netstat -nm
>>> 61785/76680 mbufs in use:
>>> 	54018 mbufs allocated to data
>>> 	7527 mbufs allocated to packet headers
>>> 	240 mbufs allocated to packet tags
>>> 	14895 mbufs allocated to caches
>>> 7991/10098 mbuf 2KB clusters in use
>>> 20468/44823 mbuf 4KB clusters in use
>>> 10909/10922 mbuf 16KB clusters in use
>>> 398566 KB allocated to network (72.9% in use)
>>> 0 KB returned to the system
>>> 0 requests for memory denied
>>> 10407 requests for memory delayed
>>> 136 calls to drain routines
>>> 
>> I wonder if increasing the memory for mbuf clusters would help. Don't try this at home but I think `sudo nvram boot-args="ncl=797132"` + reboot will double it on this system. 
>> 
>> -Alan
>> 
>> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/loom-dev/attachments/20231108/f0a231f9/attachment.htm>