- chore(go): use go 1.22.4
- use github.com/Scalingo/go-utils/graceful for graceful upgrades and shutdowns
- chore(deps): various updates
- Bump github.com/fsouza/go-dockerclient from 1.8.3 to 1.9.0
- Bump github.com/Scalingo/go-handlers from 1.4.5 to 1.5.0
- Bump github.com/tklauser/go-sysconf from 0.3.10 to 0.3.11
- Wrap all errors of the project
- Replace
errgo
package bypkg/errors
for errors handling - Don't return an error when memory can't be fetched for one container, just skip the container and return normally metrics at the end
- Add Error Middleware to the router
- Bump github.com/Scalingo/go-handlers from 1.4.4 to 1.4.5
- Bump github.com/fsouza/go-dockerclient from 1.8.1 to 1.8.3
- Bump github.com/Scalingo/go-utils/logger from 1.1.1 to 1.2.0
- Bump github.com/Scalingo/go-handlers from 1.4.3 to 1.4.4
- Bump github.com/stretchr/testify from v1.7.1 to v1.8.0
- Bump github.com/fsouza/go-dockerclient from v1.7.11 to v1.8.1
- Bump github.com/Scalingo/go-utils/logger from v1.1.0 to v1.1.1
- Bump github.com/sirupsen/logrus from 1.8.1 to 1.9.0
- chore(go): use go 1.17
- Bump various dependencies
- Bump Go version to 1.16
- Bump github.com/golang/mock from 1.3.1 to 1.5.0
- Bump github.com/stretchr/testify from 1.5.1 to 1.7.0
- Bump github.com/Scalingo/go-handlers from 1.3.1 to 1.4.0
- Bump github.com/sirupsen/logrus from 1.7.0 to 1.8.1
- Bump github.com/tklauser/go-sysconf from v0.0.0-20200513113950-67a71062da8a to 0.3.9
- Bump dependencies, stop using github.com/Scalingo/go-utils globally and uses submodules
- Add Host resources monitoring
- Add authentication
- Fix CPU value reading
- Fix int overflow when parsing cpuacct cgroup file
- Ignore first values of net/cpu monitoring to avoid extreme values (return 0 instead)
- Migration from godep to dep
- Improve logging using logrus, less spammy
- Replace martini by gorilla/mux and negroni
- Net monitoring improvement
-> Just setns/fork to find the host network interface ID, then directly
read
veth
value from the host without having to enter the namespace again
- Fix race condition
- Fix memory monitoring: negative swap values were possible before, not anymore
- Fix logic in net monitoring
- Let user disable Net monitoring
- BUGFIX fd leak when reading cgroup file
- NEW add labels to /containers/usage to get metadata of containers
- NEW /containers/usage to get cpu + mem for all containers
- /containers/:id/cpu returns an object with
.usage_in_percents
- /containers/:id/memory drops
.swap_memory_usage
,.swap_memory_limit
and.max_swap_memory
- /containers/:id/memory get
.swap_usage
,.swap_limit
and.max_swap_usage
- NEW /containers/:id/usage to get cpu + mem and optionally net usage
- Remove dependency to app from API client (client package)
- Fix nil exception when looking for memory data
- Update client package to fit new API
- Change runner PATH model
- Rename net runner to acadock-monitoring-ns-netstat
- Can run in a container
- Stats of net interface of containers
- More memory stats
- CPU monitoring "not ready" -> -1
- Parse memory on 64bits
- Metrics with Docker - Systemd
- Metrics with Docker - Libcontainer