NetSUS linux AFP service needs constant restart to remain functional

glopez1
New Contributor II

snippet from afpd.log (on server) with netbooted casper imaging client trying to mount the AFP share during imaging process:

Feb 23 11:24:26.082792 afpd[7713] {dsi_tcp.c:212} (I:DSI): AFP/TCP session from 10.45.22.189:62058
Feb 23 11:24:26.083453 afpd[1573] {main.c:183} (I:AFPDaemon): child[7712]: exited 1
Feb 23 11:24:26.084170 afpd[1573] {main.c:185} (I:AFPDaemon): child[7713]: done
Feb 23 11:24:41.219410 afpd[7714] {dsi_tcp.c:212} (I:DSI): AFP/TCP session from 10.45.22.189:62063
Feb 23 11:24:41.220504 afpd[7714] {uams_dhx2_pam.c:321} (I:UAMS): DHX2 login: afpReadOnly
Feb 23 11:24:41.300196 afpd[7714] {uams_dhx2_pam.c:210} (I:UAMS): PAM DHX2: PAM Success
Feb 23 11:24:41.377491 afpd[7714] {uams_dhx2_pam.c:653} (I:UAMS): DHX2: PAM Auth OK!
Feb 23 11:24:41.377525 afpd[7714] {auth.c:269} (N:AFPDaemon): AFP3.3 Login by afpReadOnly
Feb 23 11:24:41.382783 afpd[7714] {volume.c:2028} (I:AFPDaemon): CNID server: localhost:4700
Feb 23 11:24:41.415761 afpd[7714] {quota.c:577} (I:AFPDaemon): getquota: special /home/afpdistributionshare/CasperShare fails
Feb 23 11:27:33.412425 afpd[7714] {auth.c:948} (N:AFPDaemon): AFP logout by afpReadOnly
Feb 23 11:27:33.414132 afpd[7714] {afp_dsi.c:99} (N:AFPDaemon): AFP statistics: 1211.85 KB read, 8594089.16 KB written
Feb 23 11:27:33.414159 afpd[7714] {dircache.c:615} (I:AFPDaemon): dircache statistics: entries: 19, lookups: 624, hits: 603, misses: 21, added: 19, removed: 0, expunged: 0, evicted: 0
Feb 23 11:27:33.414198 afpd[7714] {afp_dsi.c:561} (N:AFPDaemon): done
Feb 23 11:27:33.415073 afpd[1573] {main.c:185} (I:AFPDaemon): child[7714]: done

snippet from Activity Log on netbooted client:

mount_afp: FindValidIPAddress - connect failed socket 0, optionValue 61

All it takes is restarting the afp service on our linux NetSUS and it works again. The logs show a successful connection despite the supposed "failure", and the getquota: special /home/afpdistributionshare/CasperShare fails on both successful and unsuccessful AFP connections. I've read its harmless and ignorable.

Anyone have any ideas why this is spontaneously dying after about a week and requiring a restart?

2 REPLIES 2

davidacland
Honored Contributor II

We've got a bunch of NetSUSes deployed and all are working ok. Assuming there isn't a weird bug with the particular version you are using, the only other thought is whether its running out of memory after a period of time. afpd can be fairly memory hungry.

glopez1
New Contributor II

@davidacland I appreciate your response, I will look into it. Knowing there are functional ones in the wild likely localizes the issue to me.