3 Commits

Author SHA1 Message Date
Harry-zklcdc ef93c40361 [Fix] 🐛 Fix GPU Device Error at NoCard Instances #14 2025-02-15 22:00:02 +08:00
Harry-zklcdc 716764a86d [Fix] 🐛 From Status Error at Start Action #12 2025-02-12 23:25:53 +08:00
Harry-zklcdc 44c319698c [Fix] 🐛 Mount Point Error at Container internal #10 2025-02-12 18:32:09 +08:00
2 changed files with 3 additions and 2 deletions
+1 -1
View File
@@ -61,7 +61,7 @@ func control(serverID uint, data Data) (err error) {
if err != nil {
ctx := context.Background()
redis.RawDB.IncrBy(ctx, "remain_gpu:server:"+strconv.Itoa(int(serverID)), int64(instance.GpuCount))
database.DB.Model(&instance).Update("status", models.InstanceStatusFail).Update("from_action", models.InstanceActionRestart)
database.DB.Model(&instance).Update("status", models.InstanceStatusFail).Update("from_action", models.InstanceActionStart)
lc.Error("instance restart error: %v", err)
return
}
@@ -48,6 +48,7 @@ func createInstance(ip string, port int, apikey string,
},
Env: []string{
"NVIDIA_DRIVER_CAPABILITIES=video,compute,utility",
"NVIDIA_VISIBLE_DEVICES=none",
},
}
@@ -63,7 +64,7 @@ func createInstance(ip string, port int, apikey string,
if config.GetSystemMountDir() != "" {
data.Binds = append(data.Binds, bindStruct{
Src: config.GetSystemMountDir(),
Dest: "/root/megrez",
Dest: "/root/megrez-pub",
})
}