The default value for rotate_wait in the tail input plugin does not work on s390x

[rightblank]

Bug Report

Describe the bug

The rotate_wait property of the tail input plugin on system s390x is 0 by default rather than 5 as specified in the doc

• Fluent-bit works fine on x86 with the same docker command in the Steps to reproduce the problem section
• Specify the value in the configuration file would help to avoid the issue, but it's unknown how many configs are there with the same issue.

To Reproduce

• Rubular link if applicable: None
• Example log message if applicable:

docker run -ti --rm cr.fluentbit.io/fluent/fluent-bit:3.0.3 /fluent-bit/bin/fluent-bit -i tail  -p path=/var/log/syslog -o stdout
Fluent Bit v3.0.3
* Copyright (C) 2015-2024 The Fluent Bit Authors
* Fluent Bit is a CNCF sub-project under the umbrella of Fluentd
* https://fluentbit.io/

___________.__                        __    __________.__  __          ________  
\_   _____/|  |  __ __   ____   _____/  |_  \______   \__|/  |_  ___  _\_____  \ 
 |    __)  |  | |  |  \_/ __ \ /    \   __\  |    |  _/  \   __\ \  \/ / _(__  < 
 |     \   |  |_|  |  /\  ___/|   |  \  |    |    |   \  ||  |    \   / /       \
 \___  /   |____/____/  \___  >___|  /__|    |______  /__||__|     \_/ /______  /
     \/                     \/     \/               \/                        \/ 

[2024/05/15 05:40:55] [ info] [fluent bit] version=3.0.3, commit=3529bbb132, pid=1
[2024/05/15 05:40:55] [ info] [storage] ver=1.5.2, type=memory, sync=normal, checksum=off, max_chunks_up=128
[2024/05/15 05:40:55] [ info] [cmetrics] version=0.9.0
[2024/05/15 05:40:55] [ info] [ctraces ] version=0.5.1
[2024/05/15 05:40:55] [ info] [input:tail:tail.0] initializing
[2024/05/15 05:40:55] [ info] [input:tail:tail.0] storage_strategy='memory' (memory only)
[2024/05/15 05:40:55] [error] [input:tail:tail.0] invalid 'rotate_wait' config value
[2024/05/15 05:40:55] [error] failed initialize input tail.0
[2024/05/15 05:40:55] [error] [engine] input initialization failed

• Steps to reproduce the problem:

Start the fluent-bit:3.0.3 docker image on s390x ubuntu machine with the commands below:

docker run -ti --rm cr.fluentbit.io/fluent/fluent-bit:3.0.3 /fluent-bit/bin/fluent-bit -i tail  -p path=/var/log/syslog -o stdout

Expected behavior

Screenshots

Your Environment

• Version used: 3.0.3
• Configuration: None,
• Environment name and version (e.g. Kubernetes? What version?): docker

$  docker info
Client:
 Context:    default
 Debug Mode: false
 Plugins:
  buildx: Docker Buildx (Docker Inc., v0.10.5)
  compose: Docker Compose (Docker Inc., v2.18.1)

Server:
 Containers: 7
  Running: 3
  Paused: 0
  Stopped: 4
 Images: 149
 Server Version: 20.10.21
 Storage Driver: overlay2
  Backing Filesystem: extfs
  Supports d_type: true
  Native Overlay Diff: true
  userxattr: false
 Logging Driver: json-file
 Cgroup Driver: cgroupfs
 Cgroup Version: 1
 Plugins:
  Volume: local
  Network: bridge host ipvlan macvlan null overlay
  Log: awslogs fluentd gcplogs gelf journald json-file local logentries splunk syslog
 Swarm: inactive
 Runtimes: io.containerd.runc.v2 io.containerd.runtime.v1.linux runc
 Default Runtime: runc
 Init Binary: docker-init
 containerd version: 
 runc version: 
 init version: 
 Security Options:
  apparmor
  seccomp
   Profile: default
 Kernel Version: 4.15.0-213-generic
 Operating System: Ubuntu 18.04.6 LTS
 OSType: linux
 Architecture: s390x
 CPUs: 8
 Total Memory: 15.54GiB
 Name: pok1-qz1-sr1-rk010-s20
 ID: SLPK:Q4W4:PPSB:HKYT:7GFN:TLQK:FXED:DN7G:FGKH:NHZP:RZ4N:4KWL
 Docker Root Dir: /data-vol/docker
 Debug Mode: false
 Registry: https://index.docker.io/v1/
 Labels:
 Experimental: false
 Insecure Registries:
  192.168.0.18:5000
  de.icr.io
  marina.eden.cloudlab.austin.ibm.com:5000
  127.0.0.0/8
 Live Restore Enabled: false

WARNING: No swap limit support

• Server type and version: z14

• Operating System and version:

$  uname -a
Linux pok1-qz1-sr1-rk010-s20 4.15.0-213-generic #224-Ubuntu SMP Mon Jun 19 13:29:31 UTC 2023 s390x s390x s390x GNU/Linux
$  cat /etc/lsb-release 
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=18.04
DISTRIB_CODENAME=bionic
DISTRIB_DESCRIPTION="Ubuntu 18.04.6 LTS"

• Filters and plugins: None

Additional context

[rightblank]

It turns out to be a endian issue, the below change in src/flb_config_map.c would help to fix it.

--- a/src/flb_config_map.c
+++ b/src/flb_config_map.c
@@ -649,7 +649,7 @@ int flb_config_map_set(struct mk_list *properties, struct mk_list *map, void *co
         }
         else if (m->type == FLB_CONFIG_MAP_TIME) {
             m_i_num = (int *) (base + m->offset);
-            *m_i_num = m->value.val.s_num;
+            *m_i_num = m->value.val.i_num;
         }

The root cause is that

• rotate_wait is stored as a 4 byte int number in the config_map.
• value.val.s_num gets it as a 8 bytes size_t number, this means num = num << 32 on big endian systems like s390x
• when assign new the value back to the int number *m_i_num, only the lower 4 bytes of 0 is passed to it, so it will be equal to 0.

[rightblank]

Similar endian issue happens at flb_config_map.c#L778-L787,

• plugins like systemd set the switch variables like lowercase as int,
• if flb_utils_bool(kv->val) returns 1 in the below code, execute*m_bool = ret will set the first byte of the integer to 1, the resulted integer will be 1<<24 on big endian systems as there are 3 bytes appeneded to the right.

            else if (m->type == FLB_CONFIG_MAP_BOOL) {
                m_bool = (char *) (base + m->offset);
                ret = flb_utils_bool(kv->val);
                if (ret == -1) {
                    flb_error("[config map] invalid value for boolean property '%s=%s'",
                              m->name, kv->val);
                    return -1;
                }
                *m_bool = ret;
            }