[Bug] heplify-server start successfully but heplify-server.log file not create ?

[hangocanh2303]

I'm using CentOS Linux release 7.7.1908 (Core)
I'm install heplify-server via this link: https://github.com/sipcapture/homer/wiki/homer-seven-setup#the-heplify-capture-agent
Result: heplify-server start successfully but heplify-server.log file not create

Here is heplify-server.toml file:
HEPAddr = "0.0.0.0:9060"
HEPTCPAddr = ""
HEPTLSAddr = "0.0.0.0:9060"
ESAddr = ""
ESDiscovery = false
LokiURL = ""
LokiBulk = 200
LokiTimer = 4
LokiBuffer = 100000
LokiHEPFilter = []
ForceHEPPayload = []
PromAddr = "0.0.0.0:9096"
PromTargetIP = ""
PromTargetName = ""
DBShema = "homer7"
DBDriver = "postgres"
DBAddr = "localhost:5432"
DBUser = "homer_user"
DBPass = "FT5QRbUOflpHMrWqXGUwu39r9cQ"
DBDataTable = "homer_data"
DBConfTable = "homer_config"
DBBulk = 200
DBTimer = 4
DBBuffer = 400000
DBWorker = 32
DBRotate = true
DBPartLog = "2h"
DBPartSip = "1h"
DBPartQos = "6h"
DBDropDays = 14
DBDropDaysCall = 0
DBDropDaysRegister = 0
DBDropDaysDefault = 0
DBDropOnStart = false
Dedup = false
DiscardMethod = ["OPTIONS","NOTIFY"]
AlegIDs = []
CustomHeader = []
SIPHeader = []
LogDbg = "hep,sql"
LogLvl = "info"
LogStd = false
LogSys = false
Config = "./heplify-server.toml"
ConfigHTTPAddr = ""

[negbie]

Mby some permission issue?

[hangocanh2303]

Mby some permission issue?

No, I changed permission of heplify-server folder to 777 and log file still not create

[negbie]

Does LogSys = true work?

[hangocanh2303]

Does LogSys = true work?

No, When I set LogSys = true, no log in /var/log/heplify-server.log

[negbie]

LogSys = true should write into syslog, normally under /var/log/syslog

[hangocanh2303]

LogSys = true should write into syslog, normally under /var/log/syslog
@negbie No syslog folder in /var/log

[negbie]

Red Hat family distributions (including CentOS and Fedora) use /var/log/messages and /var/log/secure

Maybe there?

[negbie]

I use for almost everything Debian and there it's /var/log/syslog

[negbie]

Since this seems to be an CentOS specific issue maybe someone else could help you out here? Let tag this issue properly.

[hangocanh2303]

I use for almost everything Debian and there it's /var/log/syslog

@negbie /var/log/messages printed some log of heplify, I will check it later.
But, I don't know why if syslog = false, heplify-server.log not create. It seem like centOS bug.

[hangocanh2303]

here is some log in /var/log/message
Feb 14 20:14:54 /opt/heplify-server/heplify-server[27016]: server.go:83: start heplify-server 1.15 with config.HeplifyServer{HEPAddr:"0.0.0.0:9060", HEPTCPAddr:"", HEPTLSAddr:"0.0.0.0:9060", ESAddr:"", ESDiscovery:false, ESUser:"", ESPass:"", LokiURL:"", LokiBulk:200, LokiTimer:4, LokiBuffer:100000, LokiHEPFilter:[]int{}, ForceHEPPayload:[]int{}, PromAddr:"0.0.0.0:9096", PromTargetIP:"", PromTargetName:"", DBShema:"homer7", DBDriver:"postgres", DBAddr:"localhost:5432", DBUser:"homer_user", DBPass:"FT5QRbUOflpHMrWqXGUwu39r9cQ", DBDataTable:"homer_data", DBConfTable:"homer_config", DBBulk:200, DBTimer:4, DBBuffer:400000, DBWorker:32, DBRotate:true, DBPartLog:"2h", DBPartIsup:"6h", DBPartSip:"1h", DBPartQos:"6h", DBDropDays:14, DBDropDaysCall:0, DBDropDaysRegister:0, DBDropDaysDefault:0, DBDropOnStart:false, Dedup:false, DiscardMethod:[]string{"OPTIONS", "NOTIFY"}, AlegIDs:[]string{}, CustomHeader:[]string{}, SIPHeader:[]string{}, LogDbg:"hep,sql", LogLvl:"debug", LogStd:false, LogSys:true, Config:"/opt/heplify-server/heplify-server.toml", ConfigHTTPAddr:"", ConfigHTTPPW:"", Version:false}
Feb 14 20:14:54 /opt/heplify-server/heplify-server[27016]: prometheus.go:38: expose metrics without or unbalanced targets
Feb 14 20:14:54 /opt/heplify-server/heplify-server[27016]: rotator.go:331: start creating tables (2020-02-14 20:14:54.864650105 +0700 +07 m=+0.018368067)
Feb 14 20:14:55 /opt/heplify-server/heplify-server[27016]: rotator.go:344: end creating tables (2020-02-14 20:14:55.544802807 +0700 +07 m=+0.698520771)
Feb 14 20:14:55 /opt/heplify-server/heplify-server[27016]: postgres.go:62: postgres connection established

[negbie]

Looks good. Maybe you or someone else will find out why the normal log file can't be written on CentOS. For me it's currently to low prio so help is wanted.

[systemcrash]

My first thoughts are SELinux.

ls -Z /opt/heplify/heplify.log

Try to touch /opt/heplify/heplify.log as root the target logfile (then chown appropriately for heplify user:group) and systemctl restart heplify. See if that helps. If so, 99% certain it's SELinux perms.

[negbie]

Thank you. I hope this helps some other user!

[ffloimair]

It's a general bug. There is another issue #360 that clearly points to the root cause of problems: A hardcoded path in the working directory.
If you install from qxip repository the binary will be in /usr/local/bin (which is ok), but you certainly don't want to log there.
Why not choose a logical default in /var/log and a configuration option to divert it elsewhere?

As it is now, logging to a file is basically unusable/inoperable (at least outside of dedicated containers).

Sure there is logs in syslog but if you want to have it separated this is currently not easily possible.

[lmangani]

As it is now, logging to a file is basically unusable/inoperable (at least outside of dedicated containers).

This is a 4 years old issue you're commenting on so I'm not sure what the value is. If there's a PR or a proposal, we're happy to hear it and implement it.

[ffloimair]

@lmangani No reason to be offended here.
The value of my comment is very straightforward:
I just wanted to cross-link to the still open issue that is 100% related in case someone (like I did) stumbles upon this while searching for the problem.

[lmangani]

Hey @ffloimair i'm absolutely not offended but I think opening a fresh issue with your comments, proposals, etc would be more useful than commenting on a very old closed one. Feel free to do so and we'll be happy to look at this again for a solution.